<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
	<id>https://wiki.socr.umich.edu/api.php?action=feedcontributions&amp;feedformat=atom&amp;user=Dinov</id>
	<title>SOCR - User contributions [en]</title>
	<link rel="self" type="application/atom+xml" href="https://wiki.socr.umich.edu/api.php?action=feedcontributions&amp;feedformat=atom&amp;user=Dinov"/>
	<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php/Special:Contributions/Dinov"/>
	<updated>2026-07-18T13:27:56Z</updated>
	<subtitle>User contributions</subtitle>
	<generator>MediaWiki 1.31.6</generator>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027&amp;diff=18389</id>
		<title>SOCR JMM 2027</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027&amp;diff=18389"/>
		<updated>2026-07-11T12:19:22Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Session Overview */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: 2027 JMM/AMS Special Session (ID: 2231) on ''Mathematical Foundations of Artificial Intelligence and Machine Learning'' ==&lt;br /&gt;
&lt;br /&gt;
[[Image:BigData_AMS_JMM_2014.gif|250px|thumbnail|right| [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-agenda 2027 JMM/AMS] ]]&lt;br /&gt;
&lt;br /&gt;
==Session Overview==&lt;br /&gt;
&lt;br /&gt;
* ''2027 AMS/JMM'': Annual [https://jointmathematicsmeetings.org/jmm 2027 JMM Congress]&lt;br /&gt;
&lt;br /&gt;
* [https://meetings.ams.org/math/jmm2027/ ''Session SS57A'': AMS Special Session (ID: 2231) on Mathematical Foundations of Artificial Intelligence and Machine Learning, I]&lt;br /&gt;
** ''Date'': Tuesday, January 12, 2027: 8:00 AM - 12:00 PM, South, '''S-501a'''&lt;br /&gt;
&lt;br /&gt;
* [https://meetings.ams.org/math/jmm2027/ ''Session SS57B'': AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, II]&lt;br /&gt;
** ''Date'': Tuesday, January 12, 2027: 1:00 PM - 5:00 PM, South, '''S-501a''' &lt;br /&gt;
&lt;br /&gt;
* ''Organizers'': [https://www.isu.edu/math/people/tenure/ Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University)] and [https://socr.umich.edu/people/dinov/ Ivo D. Dinov (University of Michigan)]&lt;br /&gt;
&lt;br /&gt;
* ''JMM-2027 Venue'': [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-info JMM 2027 takes place at the McCormick Place — Lakeside Center, 2301 S. Jean Baptiste Point DuSable Lake Shore Drive, Chicago, IL 60616]&lt;br /&gt;
&lt;br /&gt;
== Abstract Submission==&lt;br /&gt;
&lt;br /&gt;
* ''Abstract'' Submission: [https://meetings.ams.org/math/jmm2027/cfp.cgi Abstracts must be submitted through the AMS portal].&lt;br /&gt;
* ''Submission Deadline'': September 15, 2026&lt;br /&gt;
* ''Logistics'': Each speaker must submit an abstract before a talk can be approved and scheduled. Abstracts must be submitted electronically through the [https://meetings.ams.org/math/jmm2027/cfp.cgi AMS portal]. The deadline for abstract submission is midnight Eastern time on Tuesday, September 15, 2026. This is a hard deadline that cannot be extended. We strongly encourage you to submit your abstract at least a week before that deadline to avoid any last-minute problems. When the &amp;quot;Conclude Submission&amp;quot; button is clicked, an email will be sent to the Presenting Author's email and the Submitter's email confirming receipt of the submission. Please note your abstract is not submitted until you click the &amp;quot;Conclude Submission&amp;quot; button. All talks must be delivered via a computer projection system. Please note that overhead projectors, whiteboards, and blackboards will not be provided for sessions.  The AMS does not pay any expenses of presenters attending special sessions. However, graduate students are encouraged to [http://www.ams.org/programs/travel-grants/grad-students/emp-student-JMM apply for travel funding from AMS], for more information. The AMS also has a program of [https://www.ams.org/grants-awards/childcaregrants childcare grants for all JMM attendees (faculty, students, etc.)].&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
== [https://meetings.ams.org/math/jmm2027/  Session Program]==&lt;br /&gt;
&lt;br /&gt;
.. TBD ...&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18388</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18388"/>
		<updated>2026-07-11T12:18:38Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2027 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2027==&lt;br /&gt;
* January 12-15, 2027: At the annual [https://jointmathematicsmeetings.org/jmm 2027 JMM Congress], Maryam Bagherian, Emanuele Zappala, and Ivo Dinov are organizing a [[SOCR_JMM_2027 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning'']]. [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-info JMM 2027 takes place at the McCormick Place — Lakeside Center, 2301 S. Jean Baptiste Point DuSable Lake Shore Drive, Chicago, IL 60616].&lt;br /&gt;
** Ivo Dinov will be presenting a talk ''(TBD)'' at this 2027 JMM invited special session.&lt;br /&gt;
** '''Tuesday, January 12, 2027: 8:00 AM - 12:00 PM''', South, S-501a [ '''SS57A''' AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, I]&lt;br /&gt;
** '''Tuesday, January 12, 2027: 1:00 PM - 5:00 PM''', South, S-501a [ '''SS57B''' AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, II].&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 10-12, 2026: Ivo Dinov and other APS-GDS colleagues are co-organizing [https://datascience.stanford.edu/events/center-decoding-universe/c4du-annual-conference/2026-conference-physics-and-ai-pai26 2026 Conference on Physics and AI (PAI26)] at the Stanford University [https://datascience.stanford.edu/kipacsds-center-decoding-universe Center for Decoding the Universe].&lt;br /&gt;
&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
** June 1, 2026: Ivo Dinov's talk on [https://asa-sii.github.io/website/ASA-SII-2026/#schedule Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain], '''Track B · S10 Session''': ''Emerging Statistical Techniques for Longitudinal and Multimodal Imaging Analysis'', [https://wiki.socr.umich.edu/images/f/f6/ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf Dinov's KPT Slidedeck].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2026: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027&amp;diff=18387</id>
		<title>SOCR JMM 2027</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027&amp;diff=18387"/>
		<updated>2026-07-11T12:17:56Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: 2027 JMM/AMS Special Session (ID: 2231) on ''Mathematical Foundations of Artificial Intelligence and Machine Learning'' ==&lt;br /&gt;
&lt;br /&gt;
[[Image:BigData_AMS_JMM_2014.gif|250px|thumbnail|right| [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-agenda 2027 JMM/AMS] ]]&lt;br /&gt;
&lt;br /&gt;
==Session Overview==&lt;br /&gt;
&lt;br /&gt;
* ''2027 AMS/JMM'': Annual [https://jointmathematicsmeetings.org/jmm 2027 JMM Congress]&lt;br /&gt;
&lt;br /&gt;
* [https://meetings.ams.org/math/jmm2027/ ''Session SS57A'': AMS Special Session (ID: 2231) on Mathematical Foundations of Artificial Intelligence and Machine Learning, I]&lt;br /&gt;
** ''Date'': Tuesday, January 12, 2027: 8:00 AM - 12:00 PM, South, '''S-501a'''&lt;br /&gt;
&lt;br /&gt;
* [https://meetings.ams.org/math/jmm2027/ ''Session SS57B'': AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, II]&lt;br /&gt;
** ''Date'': Tuesday, January 12, 2027: 1:00 PM - 5:00 PM, South, '''S-501a''' &lt;br /&gt;
&lt;br /&gt;
* ''Organizers'': [https://www.isu.edu/math/people/tenure/ Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University)] and [https://socr.umich.edu/people/dinov/ Ivo D. Dinov (University of Michigan)]&lt;br /&gt;
&lt;br /&gt;
== Abstract Submission==&lt;br /&gt;
&lt;br /&gt;
* ''Abstract'' Submission: [https://meetings.ams.org/math/jmm2027/cfp.cgi Abstracts must be submitted through the AMS portal].&lt;br /&gt;
* ''Submission Deadline'': September 15, 2026&lt;br /&gt;
* ''Logistics'': Each speaker must submit an abstract before a talk can be approved and scheduled. Abstracts must be submitted electronically through the [https://meetings.ams.org/math/jmm2027/cfp.cgi AMS portal]. The deadline for abstract submission is midnight Eastern time on Tuesday, September 15, 2026. This is a hard deadline that cannot be extended. We strongly encourage you to submit your abstract at least a week before that deadline to avoid any last-minute problems. When the &amp;quot;Conclude Submission&amp;quot; button is clicked, an email will be sent to the Presenting Author's email and the Submitter's email confirming receipt of the submission. Please note your abstract is not submitted until you click the &amp;quot;Conclude Submission&amp;quot; button. All talks must be delivered via a computer projection system. Please note that overhead projectors, whiteboards, and blackboards will not be provided for sessions.  The AMS does not pay any expenses of presenters attending special sessions. However, graduate students are encouraged to [http://www.ams.org/programs/travel-grants/grad-students/emp-student-JMM apply for travel funding from AMS], for more information. The AMS also has a program of [https://www.ams.org/grants-awards/childcaregrants childcare grants for all JMM attendees (faculty, students, etc.)].&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
== [https://meetings.ams.org/math/jmm2027/  Session Program]==&lt;br /&gt;
&lt;br /&gt;
.. TBD ...&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027&amp;diff=18386</id>
		<title>SOCR JMM 2027</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027&amp;diff=18386"/>
		<updated>2026-07-11T12:17:17Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: 2027 JMM/AMS Special Session (ID: 2231) on ''Mathematical Foundation of Machine Learning'' ==&lt;br /&gt;
&lt;br /&gt;
[[Image:BigData_AMS_JMM_2014.gif|250px|thumbnail|right| [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-agenda 2027 JMM/AMS] ]]&lt;br /&gt;
&lt;br /&gt;
==Session Overview==&lt;br /&gt;
&lt;br /&gt;
* ''2027 AMS/JMM'': Annual [https://jointmathematicsmeetings.org/jmm 2027 JMM Congress]&lt;br /&gt;
&lt;br /&gt;
* [https://meetings.ams.org/math/jmm2027/ ''Session SS57A'': AMS Special Session (ID: 2231) on Mathematical Foundations of Artificial Intelligence and Machine Learning, I]&lt;br /&gt;
** ''Date'': Tuesday, January 12, 2027: 8:00 AM - 12:00 PM, South, '''S-501a'''&lt;br /&gt;
&lt;br /&gt;
* [https://meetings.ams.org/math/jmm2027/ ''Session SS57B'': AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, II]&lt;br /&gt;
** ''Date'': Tuesday, January 12, 2027: 1:00 PM - 5:00 PM, South, '''S-501a''' &lt;br /&gt;
&lt;br /&gt;
* ''Organizers'': [https://www.isu.edu/math/people/tenure/ Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University)] and [https://socr.umich.edu/people/dinov/ Ivo D. Dinov (University of Michigan)]&lt;br /&gt;
&lt;br /&gt;
== Abstract Submission==&lt;br /&gt;
&lt;br /&gt;
* ''Abstract'' Submission: [https://meetings.ams.org/math/jmm2027/cfp.cgi Abstracts must be submitted through the AMS portal].&lt;br /&gt;
* ''Submission Deadline'': September 15, 2026&lt;br /&gt;
* ''Logistics'': Each speaker must submit an abstract before a talk can be approved and scheduled. Abstracts must be submitted electronically through the [https://meetings.ams.org/math/jmm2027/cfp.cgi AMS portal]. The deadline for abstract submission is midnight Eastern time on Tuesday, September 15, 2026. This is a hard deadline that cannot be extended. We strongly encourage you to submit your abstract at least a week before that deadline to avoid any last-minute problems. When the &amp;quot;Conclude Submission&amp;quot; button is clicked, an email will be sent to the Presenting Author's email and the Submitter's email confirming receipt of the submission. Please note your abstract is not submitted until you click the &amp;quot;Conclude Submission&amp;quot; button. All talks must be delivered via a computer projection system. Please note that overhead projectors, whiteboards, and blackboards will not be provided for sessions.  The AMS does not pay any expenses of presenters attending special sessions. However, graduate students are encouraged to [http://www.ams.org/programs/travel-grants/grad-students/emp-student-JMM apply for travel funding from AMS], for more information. The AMS also has a program of [https://www.ams.org/grants-awards/childcaregrants childcare grants for all JMM attendees (faculty, students, etc.)].&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
== [https://meetings.ams.org/math/jmm2027/  Session Program]==&lt;br /&gt;
&lt;br /&gt;
.. TBD ...&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18385</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18385"/>
		<updated>2026-07-01T22:23:08Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2027 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2027==&lt;br /&gt;
* January 12-15, 2027: At the annual [https://jointmathematicsmeetings.org/jmm 2027 JMM Congress], Maryam Bagherian, Emanuele Zappala, and Ivo Dinov are organizing a [[SOCR_JMM_2027 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning'']]. [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-info JMM 2027 takes olace at the McCormick Place — Lakeside Center, 2301 S. Jean Baptiste Point DuSable Lake Shore Drive, Chicago, IL 60616].&lt;br /&gt;
** Ivo Dinov will be presenting a talk ''(TBD)'' at this 2027 JMM invited special session.&lt;br /&gt;
** '''Tuesday, January 12, 2027: 8:00 AM - 12:00 PM''', South, S-501a [ '''SS57A''' AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, I]&lt;br /&gt;
** '''Tuesday, January 12, 2027: 1:00 PM - 5:00 PM''', South, S-501a [ '''SS57B''' AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, II].&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 10-12, 2026: Ivo Dinov and other APS-GDS colleagues are co-organizing [https://datascience.stanford.edu/events/center-decoding-universe/c4du-annual-conference/2026-conference-physics-and-ai-pai26 2026 Conference on Physics and AI (PAI26)] at the Stanford University [https://datascience.stanford.edu/kipacsds-center-decoding-universe Center for Decoding the Universe].&lt;br /&gt;
&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
** June 1, 2026: Ivo Dinov's talk on [https://asa-sii.github.io/website/ASA-SII-2026/#schedule Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain], '''Track B · S10 Session''': ''Emerging Statistical Techniques for Longitudinal and Multimodal Imaging Analysis'', [https://wiki.socr.umich.edu/images/f/f6/ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf Dinov's KPT Slidedeck].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2026: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027&amp;diff=18384</id>
		<title>SOCR JMM 2027</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027&amp;diff=18384"/>
		<updated>2026-07-01T22:18:29Z</updated>

		<summary type="html">&lt;p&gt;Dinov: Created page with &amp;quot;==  SOCR News &amp;amp; Events: 2027 JMM/AMS Special Session on ''Mathematical Foundation of Machine Learning'' ==  Image:BigData_AMS_JMM_2014.gif|250px|thumbnail|rig...&amp;quot;&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: 2027 JMM/AMS Special Session on ''Mathematical Foundation of Machine Learning'' ==&lt;br /&gt;
&lt;br /&gt;
[[Image:BigData_AMS_JMM_2014.gif|250px|thumbnail|right| [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-agenda 2027 JMM/AMS] ]]&lt;br /&gt;
&lt;br /&gt;
==Session Overview==&lt;br /&gt;
&lt;br /&gt;
* ''2027 AMS/JMM'': Annual [https://jointmathematicsmeetings.org/jmm 2027 JMM Congress]&lt;br /&gt;
&lt;br /&gt;
* [https://meetings.ams.org/math/jmm2027/ ''Session SS57A'': AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, I]&lt;br /&gt;
** ''Date'': Tuesday, January 12, 2027: 8:00 AM - 12:00 PM, South, '''S-501a'''&lt;br /&gt;
&lt;br /&gt;
* [https://meetings.ams.org/math/jmm2027/ ''Session SS57B'': AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, II]&lt;br /&gt;
** ''Date'': Tuesday, January 12, 2027: 1:00 PM - 5:00 PM, South, '''S-501a''' &lt;br /&gt;
&lt;br /&gt;
* ''Organizers'': [https://www.isu.edu/math/people/tenure/ Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University)] and [https://socr.umich.edu/people/dinov/ Ivo D. Dinov (University of Michigan)]&lt;br /&gt;
&lt;br /&gt;
== Abstract Submission==&lt;br /&gt;
&lt;br /&gt;
* Abstract Submission: [https://meetings.ams.org/math/jmm2027/cfp.cgi Abstracts must be submitted through the AMS portal].&lt;br /&gt;
* Submission Period: ...&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
== [https://meetings.ams.org/math/jmm2027/  Session Program]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_JMM_2027}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18383</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18383"/>
		<updated>2026-07-01T22:11:53Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2027 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2027==&lt;br /&gt;
* January12-15, 2027: At the annual [https://jointmathematicsmeetings.org/jmm 2027 JMM Congress], Maryam Bagherian, Emanuele Zappala, and Ivo Dinov are organizing a [[SOCR_JMM_2027 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning'']]. [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-info JMM 2027 takes olace at the McCormick Place — Lakeside Center, 2301 S. Jean Baptiste Point DuSable Lake Shore Drive, Chicago, IL 60616].&lt;br /&gt;
** Ivo Dinov will be presenting a talk ''(TBD)'' at this 2027 JMM invited special session.&lt;br /&gt;
** '''Tuesday, January 12, 2027: 8:00 AM - 12:00 PM''', South, S-501a [ '''SS57A''' AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, I]&lt;br /&gt;
** '''Tuesday, January 12, 2027: 1:00 PM - 5:00 PM''', South, S-501a [ '''SS57B''' AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, II].&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 10-12, 2026: Ivo Dinov and other APS-GDS colleagues are co-organizing [https://datascience.stanford.edu/events/center-decoding-universe/c4du-annual-conference/2026-conference-physics-and-ai-pai26 2026 Conference on Physics and AI (PAI26)] at the Stanford University [https://datascience.stanford.edu/kipacsds-center-decoding-universe Center for Decoding the Universe].&lt;br /&gt;
&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
** June 1, 2026: Ivo Dinov's talk on [https://asa-sii.github.io/website/ASA-SII-2026/#schedule Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain], '''Track B · S10 Session''': ''Emerging Statistical Techniques for Longitudinal and Multimodal Imaging Analysis'', [https://wiki.socr.umich.edu/images/f/f6/ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf Dinov's KPT Slidedeck].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2026: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18382</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18382"/>
		<updated>2026-07-01T22:10:40Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2027==&lt;br /&gt;
* January12-15, 2027: At the annual [https://jointmathematicsmeetings.org/jmm 2027 JMM Congress], Maryam Bagherian, Emanuele Zappala, and Ivo Dinov are organizing a [[SOCR_JMM_2027 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning'']]. [https://jointmathematicsmeetings.org/meetings/national/jmm2027/jmm2027-info JMM 2027 takes olace at the McCormick Place — Lakeside Center, 2301 S. Jean Baptiste Point DuSable Lake Shore Drive, Chicago, IL 60616].&lt;br /&gt;
** Ivo Dinov will be presenting a talk ''...'' at this 2027 JMM invited special session.&lt;br /&gt;
** '''Tuesday, January 12, 2027: 8:00 AM - 12:00 PM''', South, S-501a [ '''SS57A''' AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, I]&lt;br /&gt;
** '''Tuesday, January 12, 2027: 1:00 PM - 5:00 PM''', South, S-501a [ '''SS57B''' AMS Special Session on Mathematical Foundations of Artificial Intelligence and Machine Learning, II].&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 10-12, 2026: Ivo Dinov and other APS-GDS colleagues are co-organizing [https://datascience.stanford.edu/events/center-decoding-universe/c4du-annual-conference/2026-conference-physics-and-ai-pai26 2026 Conference on Physics and AI (PAI26)] at the Stanford University [https://datascience.stanford.edu/kipacsds-center-decoding-universe Center for Decoding the Universe].&lt;br /&gt;
&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
** June 1, 2026: Ivo Dinov's talk on [https://asa-sii.github.io/website/ASA-SII-2026/#schedule Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain], '''Track B · S10 Session''': ''Emerging Statistical Techniques for Longitudinal and Multimodal Imaging Analysis'', [https://wiki.socr.umich.edu/images/f/f6/ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf Dinov's KPT Slidedeck].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2026: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18381</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18381"/>
		<updated>2026-05-06T12:40:26Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2026 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 10-12, 2026: Ivo Dinov and other APS-GDS colleagues are co-organizing [https://datascience.stanford.edu/events/center-decoding-universe/c4du-annual-conference/2026-conference-physics-and-ai-pai26 2026 Conference on Physics and AI (PAI26)] at the Stanford University [https://datascience.stanford.edu/kipacsds-center-decoding-universe Center for Decoding the Universe].&lt;br /&gt;
&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
** June 1, 2026: Ivo Dinov's talk on [https://asa-sii.github.io/website/ASA-SII-2026/#schedule Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain], '''Track B · S10 Session''': ''Emerging Statistical Techniques for Longitudinal and Multimodal Imaging Analysis'', [https://wiki.socr.umich.edu/images/f/f6/ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf Dinov's KPT Slidedeck].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2023: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18380</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18380"/>
		<updated>2026-05-06T12:40:16Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2026 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 10-12, 2026: Ivo DInov and other APS-GDS colleagues are co-organizing [https://datascience.stanford.edu/events/center-decoding-universe/c4du-annual-conference/2026-conference-physics-and-ai-pai26 2026 Conference on Physics and AI (PAI26)] at the Stanford University [https://datascience.stanford.edu/kipacsds-center-decoding-universe Center for Decoding the Universe].&lt;br /&gt;
&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
** June 1, 2026: Ivo Dinov's talk on [https://asa-sii.github.io/website/ASA-SII-2026/#schedule Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain], '''Track B · S10 Session''': ''Emerging Statistical Techniques for Longitudinal and Multimodal Imaging Analysis'', [https://wiki.socr.umich.edu/images/f/f6/ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf Dinov's KPT Slidedeck].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2023: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18379</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18379"/>
		<updated>2026-04-28T17:06:45Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2026 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
** June 1, 2026: Ivo Dinov's talk on [https://asa-sii.github.io/website/ASA-SII-2026/#schedule Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain], '''Track B · S10 Session''': ''Emerging Statistical Techniques for Longitudinal and Multimodal Imaging Analysis'', [https://wiki.socr.umich.edu/images/f/f6/ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf Dinov's KPT Slidedeck].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2023: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=File:ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf&amp;diff=18378</id>
		<title>File:ASA SMI 2026 SOCR SKA KPT V3 sm.pdf</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=File:ASA_SMI_2026_SOCR_SKA_KPT_V3_sm.pdf&amp;diff=18378"/>
		<updated>2026-04-28T17:06:07Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18377</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18377"/>
		<updated>2026-04-28T13:30:31Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2026 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
** June 1, 2026: Ivo Dinov's talk on [https://asa-sii.github.io/website/ASA-SII-2026/#schedule Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain], '''Track B · S10 Session''': ''Emerging Statistical Techniques for Longitudinal and Multimodal Imaging Analysis''.&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2023: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18376</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18376"/>
		<updated>2026-04-09T18:45:29Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /*  Scientific Methods for Health Sciences - Time Series Analysis */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
* See the [https://socr.umich.edu/DSPA2/DSPA2_notes/12_LongitudinalDataAnalysis.html '''DSPA2 Extensible Longitudinal Data Analysis Chapter'''].&lt;br /&gt;
* [https://sda.statisticalcomputing.org/ The '''SOCR Statistical Data Analyzer (SDA)''' provides live app-based data and multiple tools for interactive time-varying data analytics].&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
In practice, R's ''forecast'' package fits and reports ARIMA models with seasonality, e.g.,&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
ARIMA(p,d,q)(P,D,Q)[m]&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
These six parameters define how the model handles '''trends''', '''cycles''', and '''shocks''' in the data.&lt;br /&gt;
&lt;br /&gt;
For instance, this model ''ARIMA(2,1,1)(0,1,0)[12]'' has two parts, the '''Non-Seasonal''' component (first 3 parameters), and the '''Seasonal''' component, (the last 3 parameters) .&lt;br /&gt;
&lt;br /&gt;
* Non-Seasonal Component: ''(p, d, q)'': These parameters handle the short-term relationships between consecutive observations.&lt;br /&gt;
** '''p''' (Autoregressive - AR): The number of lag observations included in the model. A value of 2 means the model uses the two previous time points to predict the next one.&lt;br /&gt;
** '''d''' (Degree of Differencing): The number of times the raw observations are differenced to make the data stationary (removing trends). A value of 1 means the model is looking at the change between points rather than the absolute values.&lt;br /&gt;
** '''q''' (Moving Average - MA): The size of the moving average window applied to forecast errors. A value of 1 means the model accounts for the error made in the previous prediction.&lt;br /&gt;
&lt;br /&gt;
* Seasonal Component: ''(P, D, Q)'': These parameters handle repeating patterns (like the AirPassengers monthly spikes).&lt;br /&gt;
** P (Seasonal Autoregressive): Similar to ''p'', but specifically for the seasonal lags. For monthly data, P=1 would mean using the value from the same month last year to predict this month, in the output, this is 0.&lt;br /&gt;
** D (Seasonal Differencing): The number of seasonal differences. A value of 1 means the model subtracted last year's value from this year's value to remove seasonal trends.&lt;br /&gt;
** Q (Seasonal Moving Average): Similar to ''q'', but for seasonal forecast errors, in the output, this is 0.&lt;br /&gt;
** [m] (Seasonal Period): The number of periods in each season. For the AirPassengers dataset, this is 12 because the data is monthly and repeats every year.&lt;br /&gt;
&lt;br /&gt;
For a specific model: '''ARIMA(2,1,1)(0,1,0)[12]'''&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Parameter !! Value !! Meaning&lt;br /&gt;
|-&lt;br /&gt;
| p || 2 || Uses 2 lags for the AR part.&lt;br /&gt;
|-&lt;br /&gt;
| d || 1 || Applied 1st order differencing to remove the general trend.&lt;br /&gt;
|-&lt;br /&gt;
| q || 1 || Uses 1 lag for the Moving Average error correction.&lt;br /&gt;
|-&lt;br /&gt;
| P || 0 || No seasonal autoregressive terms.&lt;br /&gt;
|-&lt;br /&gt;
| D || 1 || Applied seasonal differencing (Value - Value same month last year).&lt;br /&gt;
|-&lt;br /&gt;
| Q || 0 || No seasonal moving average terms.&lt;br /&gt;
|-&lt;br /&gt;
| m || 12 || Data follows a 12-month annual cycle.&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
* '''Control Parameters''': The Stepwise and Approximation Flags, stepwise=FALSE and approximation=FALSE, force ''auto.arima'' to search more thoroughly through all possible combinations of these 6 parameters rather than using a shortcut search. This results in a better fit but takes longer to compute.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines: Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
# dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Next, let's try to fit '''manually''' another (specific) '''ARIMA(1,1,5)''' model where the (p,d,q) parameters are hard-coded. Then, compare ''fitManual'' to the optimized model (''fit &amp;lt;- auto.arima()''), by contrasting the corresponding ''AIC'', ''BIC'', and log-likelihood estimates of each model, smaller values indicate better model and higher model fidelity. &lt;br /&gt;
&lt;br /&gt;
Also, compare the 24-month forward forecasts of ''fitManual'' vs. the optimized model ''fit''.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
fitManual &amp;lt;- arima(AirPassengers, order=c(1,1,5)); summary(fitManual)&lt;br /&gt;
&lt;br /&gt;
fcManual &amp;lt;- forecast(fitManual, h=24)&lt;br /&gt;
plot(fcManual, main=&amp;quot;(Manual) 24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
fcManual$model$aic  # [1] 1353.116&lt;br /&gt;
fcManual$model$loglik  # [1] -669.5579&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
&lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) &lt;br /&gt;
&lt;br /&gt;
(a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18375</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18375"/>
		<updated>2026-04-08T14:12:23Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* ARIMA Models */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
In practice, R's ''forecast'' package fits and reports ARIMA models with seasonality, e.g.,&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
ARIMA(p,d,q)(P,D,Q)[m]&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
These six parameters define how the model handles '''trends''', '''cycles''', and '''shocks''' in the data.&lt;br /&gt;
&lt;br /&gt;
For instance, this model ''ARIMA(2,1,1)(0,1,0)[12]'' has two parts, the '''Non-Seasonal''' component (first 3 parameters), and the '''Seasonal''' component, (the last 3 parameters) .&lt;br /&gt;
&lt;br /&gt;
* Non-Seasonal Component: ''(p, d, q)'': These parameters handle the short-term relationships between consecutive observations.&lt;br /&gt;
** '''p''' (Autoregressive - AR): The number of lag observations included in the model. A value of 2 means the model uses the two previous time points to predict the next one.&lt;br /&gt;
** '''d''' (Degree of Differencing): The number of times the raw observations are differenced to make the data stationary (removing trends). A value of 1 means the model is looking at the change between points rather than the absolute values.&lt;br /&gt;
** '''q''' (Moving Average - MA): The size of the moving average window applied to forecast errors. A value of 1 means the model accounts for the error made in the previous prediction.&lt;br /&gt;
&lt;br /&gt;
* Seasonal Component: ''(P, D, Q)'': These parameters handle repeating patterns (like the AirPassengers monthly spikes).&lt;br /&gt;
** P (Seasonal Autoregressive): Similar to ''p'', but specifically for the seasonal lags. For monthly data, P=1 would mean using the value from the same month last year to predict this month, in the output, this is 0.&lt;br /&gt;
** D (Seasonal Differencing): The number of seasonal differences. A value of 1 means the model subtracted last year's value from this year's value to remove seasonal trends.&lt;br /&gt;
** Q (Seasonal Moving Average): Similar to ''q'', but for seasonal forecast errors, in the output, this is 0.&lt;br /&gt;
** [m] (Seasonal Period): The number of periods in each season. For the AirPassengers dataset, this is 12 because the data is monthly and repeats every year.&lt;br /&gt;
&lt;br /&gt;
For a specific model: '''ARIMA(2,1,1)(0,1,0)[12]'''&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Parameter !! Value !! Meaning&lt;br /&gt;
|-&lt;br /&gt;
| p || 2 || Uses 2 lags for the AR part.&lt;br /&gt;
|-&lt;br /&gt;
| d || 1 || Applied 1st order differencing to remove the general trend.&lt;br /&gt;
|-&lt;br /&gt;
| q || 1 || Uses 1 lag for the Moving Average error correction.&lt;br /&gt;
|-&lt;br /&gt;
| P || 0 || No seasonal autoregressive terms.&lt;br /&gt;
|-&lt;br /&gt;
| D || 1 || Applied seasonal differencing (Value - Value same month last year).&lt;br /&gt;
|-&lt;br /&gt;
| Q || 0 || No seasonal moving average terms.&lt;br /&gt;
|-&lt;br /&gt;
| m || 12 || Data follows a 12-month annual cycle.&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
* '''Control Parameters''': The Stepwise and Approximation Flags, stepwise=FALSE and approximation=FALSE, force ''auto.arima'' to search more thoroughly through all possible combinations of these 6 parameters rather than using a shortcut search. This results in a better fit but takes longer to compute.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines: Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
# dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Next, let's try to fit '''manually''' another (specific) '''ARIMA(1,1,5)''' model where the (p,d,q) parameters are hard-coded. Then, compare ''fitManual'' to the optimized model (''fit &amp;lt;- auto.arima()''), by contrasting the corresponding ''AIC'', ''BIC'', and log-likelihood estimates of each model, smaller values indicate better model and higher model fidelity. &lt;br /&gt;
&lt;br /&gt;
Also, compare the 24-month forward forecasts of ''fitManual'' vs. the optimized model ''fit''.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
fitManual &amp;lt;- arima(AirPassengers, order=c(1,1,5)); summary(fitManual)&lt;br /&gt;
&lt;br /&gt;
fcManual &amp;lt;- forecast(fitManual, h=24)&lt;br /&gt;
plot(fcManual, main=&amp;quot;(Manual) 24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
fcManual$model$aic  # [1] 1353.116&lt;br /&gt;
fcManual$model$loglik  # [1] -669.5579&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
&lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) &lt;br /&gt;
&lt;br /&gt;
(a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18374</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18374"/>
		<updated>2026-04-08T14:12:03Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* ARIMA Models */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
In practice, R's ''forecast'' package fits and reports ARIMA models with seasonality, e.g.,&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
ARIMA(p,d,q)(P,D,Q)[m]&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
These six parameters define how the model handles '''trends''', '''cycles''', and '''shocks''' in the data.&lt;br /&gt;
&lt;br /&gt;
For instance, this model ''ARIMA(2,1,1)(0,1,0)[12]'' has two parts, the '''Non-Seasonal''' component (first 3 parameters), and the '''Seasonal''' component, (the last 3 parameters) .&lt;br /&gt;
&lt;br /&gt;
* Non-Seasonal Component: ''(p, d, q)'': These parameters handle the short-term relationships between consecutive observations.&lt;br /&gt;
&lt;br /&gt;
** '''p''' (Autoregressive - AR): The number of lag observations included in the model. A value of 2 means the model uses the two previous time points to predict the next one.&lt;br /&gt;
** '''d''' (Degree of Differencing): The number of times the raw observations are differenced to make the data stationary (removing trends). A value of 1 means the model is looking at the change between points rather than the absolute values.&lt;br /&gt;
** '''q''' (Moving Average - MA): The size of the moving average window applied to forecast errors. A value of 1 means the model accounts for the error made in the previous prediction.&lt;br /&gt;
&lt;br /&gt;
* Seasonal Component: ''(P, D, Q)'': These parameters handle repeating patterns (like the AirPassengers monthly spikes).&lt;br /&gt;
** P (Seasonal Autoregressive): Similar to ''p'', but specifically for the seasonal lags. For monthly data, P=1 would mean using the value from the same month last year to predict this month, in the output, this is 0.&lt;br /&gt;
** D (Seasonal Differencing): The number of seasonal differences. A value of 1 means the model subtracted last year's value from this year's value to remove seasonal trends.&lt;br /&gt;
** Q (Seasonal Moving Average): Similar to ''q'', but for seasonal forecast errors, in the output, this is 0.&lt;br /&gt;
** [m] (Seasonal Period): The number of periods in each season. For the AirPassengers dataset, this is 12 because the data is monthly and repeats every year.&lt;br /&gt;
&lt;br /&gt;
For a specific model: '''ARIMA(2,1,1)(0,1,0)[12]'''&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Parameter !! Value !! Meaning&lt;br /&gt;
|-&lt;br /&gt;
| p || 2 || Uses 2 lags for the AR part.&lt;br /&gt;
|-&lt;br /&gt;
| d || 1 || Applied 1st order differencing to remove the general trend.&lt;br /&gt;
|-&lt;br /&gt;
| q || 1 || Uses 1 lag for the Moving Average error correction.&lt;br /&gt;
|-&lt;br /&gt;
| P || 0 || No seasonal autoregressive terms.&lt;br /&gt;
|-&lt;br /&gt;
| D || 1 || Applied seasonal differencing (Value - Value same month last year).&lt;br /&gt;
|-&lt;br /&gt;
| Q || 0 || No seasonal moving average terms.&lt;br /&gt;
|-&lt;br /&gt;
| m || 12 || Data follows a 12-month annual cycle.&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
* '''Control Parameters''': The Stepwise and Approximation Flags, stepwise=FALSE and approximation=FALSE, force ''auto.arima'' to search more thoroughly through all possible combinations of these 6 parameters rather than using a shortcut search. This results in a better fit but takes longer to compute.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines: Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
# dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Next, let's try to fit '''manually''' another (specific) '''ARIMA(1,1,5)''' model where the (p,d,q) parameters are hard-coded. Then, compare ''fitManual'' to the optimized model (''fit &amp;lt;- auto.arima()''), by contrasting the corresponding ''AIC'', ''BIC'', and log-likelihood estimates of each model, smaller values indicate better model and higher model fidelity. &lt;br /&gt;
&lt;br /&gt;
Also, compare the 24-month forward forecasts of ''fitManual'' vs. the optimized model ''fit''.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
fitManual &amp;lt;- arima(AirPassengers, order=c(1,1,5)); summary(fitManual)&lt;br /&gt;
&lt;br /&gt;
fcManual &amp;lt;- forecast(fitManual, h=24)&lt;br /&gt;
plot(fcManual, main=&amp;quot;(Manual) 24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
fcManual$model$aic  # [1] 1353.116&lt;br /&gt;
fcManual$model$loglik  # [1] -669.5579&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
&lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) &lt;br /&gt;
&lt;br /&gt;
(a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18373</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18373"/>
		<updated>2026-04-08T13:59:12Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Modern ARIMA Modeling (Box-Jenkins Approach) */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines: Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
# dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Next, let's try to fit '''manually''' another (specific) '''ARIMA(1,1,5)''' model where the (p,d,q) parameters are hard-coded. Then, compare ''fitManual'' to the optimized model (''fit &amp;lt;- auto.arima()''), by contrasting the corresponding ''AIC'', ''BIC'', and log-likelihood estimates of each model, smaller values indicate better model and higher model fidelity. &lt;br /&gt;
&lt;br /&gt;
Also, compare the 24-month forward forecasts of ''fitManual'' vs. the optimized model ''fit''.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
fitManual &amp;lt;- arima(AirPassengers, order=c(1,1,5)); summary(fitManual)&lt;br /&gt;
&lt;br /&gt;
fcManual &amp;lt;- forecast(fitManual, h=24)&lt;br /&gt;
plot(fcManual, main=&amp;quot;(Manual) 24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
fcManual$model$aic  # [1] 1353.116&lt;br /&gt;
fcManual$model$loglik  # [1] -669.5579&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
&lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) &lt;br /&gt;
&lt;br /&gt;
(a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18372</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18372"/>
		<updated>2026-04-07T16:08:40Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Modern ARIMA Modeling (Box-Jenkins Approach) */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines: Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Next, let's try to fit '''manually''' another (specific) '''ARIMA(1,1,5)''' model where the (p,d,q) parameters are hard-coded. Then, compare ''fitManual'' to the optimized model (''fit &amp;lt;- auto.arima()''), by contrasting the corresponding ''AIC'', ''BIC'', and log-likelihood estimates of each model, smaller values indicate better model and higher model fidelity. &lt;br /&gt;
&lt;br /&gt;
Also, compare the 24-month forward forecasts of ''fitManual'' vs. the optimized model ''fit''.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
fitManual &amp;lt;- arima(AirPassengers, order=c(1,1,5)); summary(fitManual)&lt;br /&gt;
&lt;br /&gt;
fcManual &amp;lt;- forecast(fitManual, h=24)&lt;br /&gt;
plot(fcManual, main=&amp;quot;(Manual) 24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
fcManual$model$aic  # [1] 1353.116&lt;br /&gt;
fcManual$model$loglik  # [1] -669.5579&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
&lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) &lt;br /&gt;
&lt;br /&gt;
(a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18371</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18371"/>
		<updated>2026-04-07T15:13:10Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Modeling Seasonality and Trend= */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines: Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
&lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) &lt;br /&gt;
&lt;br /&gt;
(a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18370</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18370"/>
		<updated>2026-04-07T15:12:49Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Nonparametric Smoothing */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines: Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend=====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
&lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) &lt;br /&gt;
&lt;br /&gt;
(a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18369</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18369"/>
		<updated>2026-04-02T19:16:03Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Problems */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines:** Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend=====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
&lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) &lt;br /&gt;
&lt;br /&gt;
(a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18368</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18368"/>
		<updated>2026-04-02T19:14:38Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Modern ARIMA Modeling (Box-Jenkins Approach) */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines:** Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend=====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The forecast package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) (a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18367</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18367"/>
		<updated>2026-04-02T19:14:03Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Modern ARIMA Modeling (Box-Jenkins Approach) */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines:** Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend=====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The `forecast` package in `R` automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
dev.off()  # reset display, if necessary&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) (a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18366</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18366"/>
		<updated>2026-04-02T19:11:23Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Modern ARIMA Modeling (Box-Jenkins Approach) */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines:** Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend=====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. Identification: Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
2. Estimation: Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
&lt;br /&gt;
3. Diagnostic Checking: Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The `forecast` package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) (a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18365</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18365"/>
		<updated>2026-04-02T19:09:08Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines:** Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend=====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. **Identification:** Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. **Estimation:** Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
3. **Diagnostic Checking:** Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The `forecast` package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) (a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18364</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18364"/>
		<updated>2026-04-02T17:18:31Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines, such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines:** Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend=====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. **Identification:** Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. **Estimation:** Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
3. **Diagnostic Checking:** Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The `forecast` package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) (a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18363</id>
		<title>SMHS TimeSeries</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries&amp;diff=18363"/>
		<updated>2026-04-02T17:17:33Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Time Series Analysis ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Time series data is a sequence of data points measured at successive, equally spaced points in time. Time series analysis is utilized across various disciplines—such as monitoring industrial processes, tracking business metrics, and analyzing longitudinal health records. In this module, we present a comprehensive introduction to time series data and explore both classical and modern techniques used in this rapidly growing field. The goal is to extract meaningful statistics, identify underlying patterns (such as trends and seasonality), and construct predictive models. We will illustrate the application of these techniques using interactive examples in R.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
Economic data (e.g., daily share prices, monthly sales, annual income) and biophysical data (e.g., daily temperature, ECG readings, epidemiological counts) are all examples of time series data. A time series is an ordered sequence of values of a variable measured at equally spaced time intervals. What are the most effective ways to model time series data? How can we extract actionable information, make inferences about underlying mechanisms, and forecast future observations? Answering these questions requires understanding the mathematical properties of time series processes and applying robust statistical modeling techniques.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====Components of Time Series====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; can generally be decomposed into three primary components, plus an irregular error term:&lt;br /&gt;
* '''Trend Component (&amp;lt;math&amp;gt;T_t&amp;lt;/math&amp;gt;):''' A long-run increase or decrease over time (overall upward or downward movement). The trend can be linear or non-linear.&lt;br /&gt;
* '''Seasonal Component (&amp;lt;math&amp;gt;S_t&amp;lt;/math&amp;gt;):''' Short-term, regular wave-like patterns completing a cycle within a fixed period (e.g., daily, monthly, quarterly).&lt;br /&gt;
* '''Cyclical Component (&amp;lt;math&amp;gt;C_t&amp;lt;/math&amp;gt;):''' Long-term wave-like patterns with variable lengths, often measured peak-to-peak or trough-to-trough, typically tied to economic or biological cycles.&lt;br /&gt;
* '''Irregular Component (&amp;lt;math&amp;gt;I_t&amp;lt;/math&amp;gt;):''' Unpredictable, random residuals, or &amp;quot;noise,&amp;quot; which may be due to random variations or unusual events.&lt;br /&gt;
&lt;br /&gt;
These components can combine additively or multiplicatively:&lt;br /&gt;
* '''Additive Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t}+S_{t}+C_{t}+I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
* '''Multiplicative Model:''' &amp;lt;math&amp;gt;x_{t}=T_{t} \times S_{t} \times C_{t} \times I_{t}&amp;lt;/math&amp;gt;&lt;br /&gt;
A logarithmic transformation can convert a multiplicative model into an additive one: &amp;lt;math&amp;gt;\log(x_{t})=\log(T_{t})+\log(S_{t})+\log(C_{t})+\log(I_{t})&amp;lt;/math&amp;gt;.&lt;br /&gt;
* '''Probabilistic models''': The components still add or multiply together across time.&lt;br /&gt;
[[Image:SMHS Fig 1 Times Series Analysis.png|400px]]&lt;br /&gt;
&lt;br /&gt;
====Stationarity====&lt;br /&gt;
Most statistical forecasting methods assume that a time series is approximately stationary. A time series is said to be '''strictly stationary''' if its statistical properties are unaffected by a shift in time. In practice, we rely on '''weak stationarity''', which requires:&lt;br /&gt;
1. The mean is constant over time: &amp;lt;math&amp;gt;E[x_t] = \mu&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. The variance is finite and constant: &amp;lt;math&amp;gt;Var(x_t) = \gamma(0) &amp;lt; \infty&amp;lt;/math&amp;gt;.&lt;br /&gt;
3. The autocovariance between any two points depends only on the time lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; between them: &amp;lt;math&amp;gt;Cov(x_t, x_{t+h}) = \gamma(h)&amp;lt;/math&amp;gt; for all &amp;lt;math&amp;gt;t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
Real-world time series are usually non-stationary (exhibiting trends or varying variance) but can often be transformed to achieve stationarity (e.g., differencing, log transformations).&lt;br /&gt;
&lt;br /&gt;
====Autocorrelation and Partial Autocorrelation====&lt;br /&gt;
To characterize a time series, we examine its mean and covariance structure. For a stationary process, the '''Autocovariance Function (ACVF)''' at lag &amp;lt;math&amp;gt;h&amp;lt;/math&amp;gt; is defined as:&lt;br /&gt;
&amp;lt;math&amp;gt;\gamma(h) = E[(x_t - \mu)(x_{t+h} - \mu)]&amp;lt;/math&amp;gt;&lt;br /&gt;
The '''Autocorrelation Function (ACF)''' standardizes this by the variance &amp;lt;math&amp;gt;\gamma(0)&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \frac{\gamma(h)}{\gamma(0)}&amp;lt;/math&amp;gt;&lt;br /&gt;
Properties of the ACF include: &amp;lt;math&amp;gt;\rho(0) = 1&amp;lt;/math&amp;gt;, &amp;lt;math&amp;gt;|\rho(h)| \le 1&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(h) = \rho(-h)&amp;lt;/math&amp;gt; (it is an even function).&lt;br /&gt;
&lt;br /&gt;
The '''Partial Autocorrelation Function (PACF)''' measures the correlation between &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;x_{t+h}&amp;lt;/math&amp;gt; after removing the linear dependence on the intermediate observations &amp;lt;math&amp;gt;x_{t+1}, \dots, x_{t+h-1}&amp;lt;/math&amp;gt;. The ACF and PACF are crucial tools for identifying the order of Autoregressive (AR) and Moving Average (MA) models.&lt;br /&gt;
&lt;br /&gt;
=====Estimation of ACF=====&lt;br /&gt;
Given &amp;lt;math&amp;gt;T&amp;lt;/math&amp;gt; observations, the sample autocovariance is estimated as:&lt;br /&gt;
&amp;lt;math&amp;gt;\hat{\gamma}(h) = \frac{1}{T} \sum_{t=1}^{T-h} (x_t - \bar{x})(x_{t+h} - \bar{x})&amp;lt;/math&amp;gt;&lt;br /&gt;
The sample ACF is &amp;lt;math&amp;gt;\hat{\rho}(h) = \hat{\gamma}(h) / \hat{\gamma}(0)&amp;lt;/math&amp;gt;. &lt;br /&gt;
Under the null hypothesis that the true process is White Noise, the large-sample standard error of the sample ACF is approximately:&lt;br /&gt;
&amp;lt;math&amp;gt;SE(\hat{\rho}(h)) \approx \frac{1}{\sqrt{T}}&amp;lt;/math&amp;gt;&lt;br /&gt;
A common rule of thumb is that 95% of the sample autocorrelations for a White Noise process should fall within the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{T}&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====White Noise====&lt;br /&gt;
A time series &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is White Noise (&amp;lt;math&amp;gt;WN&amp;lt;/math&amp;gt;) if it is weakly stationary with:&lt;br /&gt;
&amp;lt;math&amp;gt;\rho(h) = \begin{cases} 1 &amp;amp; \text{if } h=0 \\ 0 &amp;amp; \text{if } h \neq 0 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
A special case is Gaussian White Noise, denoted &amp;lt;math&amp;gt;w_t \sim \text{WN}(0, \sigma^2)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_t&amp;lt;/math&amp;gt; is independently and identically distributed (i.i.d.) according to a normal distribution.&lt;br /&gt;
&lt;br /&gt;
====Backshift Notation and Linear Operators====&lt;br /&gt;
Modern time series theory heavily utilizes the backshift operator &amp;lt;math&amp;gt;B&amp;lt;/math&amp;gt;, defined as &amp;lt;math&amp;gt;Bx_t = x_{t-1}&amp;lt;/math&amp;gt;. &lt;br /&gt;
* Differencing can be written as &amp;lt;math&amp;gt;\nabla x_t = (1-B)x_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
* An AR(1) model &amp;lt;math&amp;gt;x_t = \phi x_{t-1} + w_t&amp;lt;/math&amp;gt; is written as &amp;lt;math&amp;gt;(1-\phi B)x_t = w_t&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Autoregressive (AR) Models====&lt;br /&gt;
An autoregressive model of order &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;, denoted AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;), models the current value as a linear combination of its past values plus a shock:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = \phi_1 x_{t-1} + \phi_2 x_{t-2} + \dots + \phi_p x_{t-p} + w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;\phi(B)x_t = w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\phi(B) = 1 - \phi_1 B - \dots - \phi_p B^p&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' The process must be stationary, which requires that the roots of &amp;lt;math&amp;gt;\phi(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. &lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of a stationary AR(&amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;) process decays gradually (tails off), while the PACF cuts off abruptly after lag &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
====Moving Average (MA) Models====&lt;br /&gt;
A moving average model of order &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, denoted MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;), models the current value as a linear combination of current and past white noise shocks:&lt;br /&gt;
&amp;lt;math&amp;gt;x_t = w_t + \theta_1 w_{t-1} + \theta_2 w_{t-2} + \dots + \theta_q w_{t-q}&amp;lt;/math&amp;gt;&lt;br /&gt;
Using backshift notation: &amp;lt;math&amp;gt;x_t = \theta(B)w_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\theta(B) = 1 + \theta_1 B + \dots + \theta_q B^q&amp;lt;/math&amp;gt;.&lt;br /&gt;
'''Assumptions:''' MA processes are always stationary. However, they must be '''invertible''', meaning the roots of &amp;lt;math&amp;gt;\theta(B) = 0&amp;lt;/math&amp;gt; lie strictly outside the unit circle. Invertibility ensures the process can be represented as an infinite-order AR model.&lt;br /&gt;
'''ACF/PACF Patterns:''' The ACF of an MA(&amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;) process cuts off abruptly after lag &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;, while the PACF decays gradually (tails off).&lt;br /&gt;
&lt;br /&gt;
====ARIMA Models====&lt;br /&gt;
The Autoregressive Integrated Moving Average model, ARIMA(&amp;lt;math&amp;gt;p, d, q&amp;lt;/math&amp;gt;), combines AR and MA models after differencing the data &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt; times to achieve stationarity:&lt;br /&gt;
&amp;lt;math&amp;gt;\phi(B)(1-B)^d x_t = \theta(B)w_t&amp;lt;/math&amp;gt;&lt;br /&gt;
If seasonality is present, we extend this to the Seasonal ARIMA (SARIMA) model, which incorporates seasonal AR, MA, and differencing terms.&lt;br /&gt;
&lt;br /&gt;
===Methods and Data Analysis===&lt;br /&gt;
&lt;br /&gt;
====Exploratory Data Analysis (EDA) and Diagnostics====&lt;br /&gt;
Before modeling, we must visualize the data, check for outliers, and assess distributional assumptions. A standard approach involves time series plots, Q-Q plots (to check normality), and examining residuals. &lt;br /&gt;
&lt;br /&gt;
If we fit a simple mean model &amp;lt;math&amp;gt;x_t = \mu + w_t&amp;lt;/math&amp;gt;, the residuals &amp;lt;math&amp;gt;e_t = x_t - \hat{\mu}&amp;lt;/math&amp;gt; should resemble White Noise. We diagnose this by checking:&lt;br /&gt;
1. Zero mean of residuals.&lt;br /&gt;
2. Constant variance across time (homoscedasticity).&lt;br /&gt;
3. No autocorrelation (checked via ACF plots of residuals).&lt;br /&gt;
4. Normality of residuals (checked via Q-Q plots).&lt;br /&gt;
&lt;br /&gt;
====Smoothing Techniques====&lt;br /&gt;
Smoothing helps identify underlying structures by averaging out the noise.&lt;br /&gt;
&lt;br /&gt;
=====Moving Averages=====&lt;br /&gt;
A simple centered moving average smooths &amp;lt;math&amp;gt;x_t&amp;lt;/math&amp;gt; using a window of size &amp;lt;math&amp;gt;k&amp;lt;/math&amp;gt;:&lt;br /&gt;
&amp;lt;math&amp;gt;v_t = \frac{1}{k} \sum_{j=-(k-1)/2}^{(k-1)/2} x_{t+j}&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Simulate White Noise and apply Moving Averages&lt;br /&gt;
set.seed(123)&lt;br /&gt;
w &amp;lt;- ts(rnorm(150))  # generate 150 data points from standard normal distribution&lt;br /&gt;
ma3 &amp;lt;- filter(w, sides=2, rep(1/3, 3))  # 3-period centered moving average&lt;br /&gt;
ma9 &amp;lt;- filter(w, sides=2, rep(1/9, 9))  # 9-period centered moving average&lt;br /&gt;
&lt;br /&gt;
plot.ts(w, main=&amp;quot;White Noise vs. Moving Averages&amp;quot;, ylab=&amp;quot;Value&amp;quot;)&lt;br /&gt;
lines(ma3, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(ma9, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
legend(&amp;quot;topright&amp;quot;, legend=c(&amp;quot;WN&amp;quot;, &amp;quot;MA(3)&amp;quot;, &amp;quot;MA(9)&amp;quot;), col=c(&amp;quot;black&amp;quot;, &amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
*Note: A larger window (lag) results in a smoother time series plot, but introduces more lag in detecting changes.*&lt;br /&gt;
&lt;br /&gt;
=====Exponential Smoothing=====&lt;br /&gt;
Unlike moving averages, exponential smoothing assigns exponentially decreasing weights as observations get older. &lt;br /&gt;
* '''Simple Exponential Smoothing (SES):''' Appropriate for data with no trend or seasonality. The forecast is &amp;lt;math&amp;gt;\hat{x}_{t+1} = \alpha x_t + (1-\alpha)\hat{x}_t&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;0 &amp;lt; \alpha &amp;lt; 1&amp;lt;/math&amp;gt; is the smoothing parameter.&lt;br /&gt;
* '''Holt-Winters Method:''' Extends SES to capture trends (additive or multiplicative) and seasonality.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Using built-in AirPassengers dataset&lt;br /&gt;
data(AirPassengers)&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;AirPassengers Dataset&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Holt-Winters Additive Model&lt;br /&gt;
hw_add &amp;lt;- HoltWinters(AirPassengers, seasonal=&amp;quot;additive&amp;quot;)&lt;br /&gt;
plot(hw_add, main=&amp;quot;Holt-Winters Additive Filtering&amp;quot;)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Nonparametric Smoothing=====&lt;br /&gt;
* '''Loess (Local Polynomial Regression):''' Fits simple models to localized subsets of data. Controlled by the `span` parameter. A larger span yields a smoother curve.&lt;br /&gt;
* '''Splines:** Piecewise polynomials joined smoothly at &amp;quot;knots.&amp;quot; Controlled by degrees of freedom or a smoothing parameter (`spar`).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Loess and Spline Smoothing on a time series&lt;br /&gt;
t &amp;lt;- time(AirPassengers)&lt;br /&gt;
fit_loess &amp;lt;- lowess(t, AirPassengers, f=0.3) # f is the span&lt;br /&gt;
fit_spline &amp;lt;- smooth.spline(t, AirPassengers, spar=0.8)&lt;br /&gt;
&lt;br /&gt;
plot(AirPassengers, main=&amp;quot;Nonparametric Smoothing&amp;quot;, ylab=&amp;quot;Passengers&amp;quot;)&lt;br /&gt;
lines(fit_loess, col=&amp;quot;red&amp;quot;, lwd=2)&lt;br /&gt;
lines(fit_spline, col=&amp;quot;blue&amp;quot;, lwd=2, lty=2)&lt;br /&gt;
legend(&amp;quot;topleft&amp;quot;, legend=c(&amp;quot;Loess&amp;quot;, &amp;quot;Spline&amp;quot;), col=c(&amp;quot;red&amp;quot;, &amp;quot;blue&amp;quot;), lwd=2, lty=c(1,2))&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Modeling Seasonality and Trend=====&lt;br /&gt;
&lt;br /&gt;
=====Classical Decomposition=====&lt;br /&gt;
We can decompose a time series into its trend, seasonal, and irregular components using moving averages or Local Regression (LOESS).&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Multiplicative decomposition using LOESS&lt;br /&gt;
decomp &amp;lt;- stl(AirPassengers, s.window=&amp;quot;periodic&amp;quot;)&lt;br /&gt;
plot(decomp, main=&amp;quot;Classical Decomposition via STL&amp;quot;)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Regression with Dummy Variables=====&lt;br /&gt;
Seasonality can be explicitly modeled using regression with dummy variables. For monthly data with an intercept, we use 11 dummy variables to avoid the dummy variable trap.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Regression with seasonal dummy variables&lt;br /&gt;
y &amp;lt;- log(AirPassengers) # Log transform to stabilize variance&lt;br /&gt;
t &amp;lt;- time(y)&lt;br /&gt;
Q &amp;lt;- factor(cycle(y))   # Creates monthly factors (1-12)&lt;br /&gt;
&lt;br /&gt;
# Fit linear regression with trend and seasonality&lt;br /&gt;
reg_model &amp;lt;- lm(y ~ t + Q)&lt;br /&gt;
summary(reg_model)&lt;br /&gt;
&lt;br /&gt;
# Diagnostics of the regression model&lt;br /&gt;
par(mfrow=c(2,2))&lt;br /&gt;
plot(reg_model, main=&amp;quot;Regression Diagnostics&amp;quot;)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
=====Modern ARIMA Modeling (Box-Jenkins Approach)=====&lt;br /&gt;
The modern standard for time series forecasting involves the Box-Jenkins iterative cycle: Identification, Estimation, and Diagnostic Checking.&lt;br /&gt;
&lt;br /&gt;
1. **Identification:** Use ACF and PACF plots, and unit root tests (like Augmented Dickey-Fuller) to determine &amp;lt;math&amp;gt;d&amp;lt;/math&amp;gt;. Use algorithms to suggest &amp;lt;math&amp;gt;p&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;q&amp;lt;/math&amp;gt;.&lt;br /&gt;
2. **Estimation:** Maximize the likelihood function to estimate AR and MA coefficients.&lt;br /&gt;
3. **Diagnostic Checking:** Ensure residuals are White Noise.&lt;br /&gt;
&lt;br /&gt;
The `forecast` package in R automates this efficiently:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Install and load forecast package if necessary&lt;br /&gt;
# install.packages(&amp;quot;forecast&amp;quot;)&lt;br /&gt;
library(forecast)&lt;br /&gt;
&lt;br /&gt;
# auto.arima automatically identifies the best SARIMA model&lt;br /&gt;
fit &amp;lt;- auto.arima(AirPassengers, seasonal=TRUE, stepwise=FALSE, approximation=FALSE)&lt;br /&gt;
summary(fit)&lt;br /&gt;
&lt;br /&gt;
# Check residuals to ensure they are White Noise&lt;br /&gt;
checkresiduals(fit)&lt;br /&gt;
&lt;br /&gt;
# Forecast the next 24 months&lt;br /&gt;
fc &amp;lt;- forecast(fit, h=24)&lt;br /&gt;
plot(fc, main=&amp;quot;24-Month Forecast using SARIMA&amp;quot;)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Outlier Detection====&lt;br /&gt;
Outliers can severely distort time series models (inflating variance, skewing ACF). They can be detected visually via time plots and boxplots, or systematically using residuals.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;source lang=&amp;quot;R&amp;quot;&amp;gt;&lt;br /&gt;
# Creating a dataset with injected outliers&lt;br /&gt;
set.seed(42)&lt;br /&gt;
dat &amp;lt;- ts(c(28,22,36,26,28,28,26,24,32,30,27,24,33,21,36,32,31,25,24,25,28,36,27,32,34,30,&lt;br /&gt;
            25,26,26,25,-44,23,21,30,33,29,27,29,28,22,26,27,16,31,29,36,32,28,40,19,&lt;br /&gt;
            37,23,32,29,-2,24,25,27,24,16,29,20,28,27,39,23))&lt;br /&gt;
&lt;br /&gt;
par(mfrow=c(1,2))&lt;br /&gt;
plot.ts(dat, main=&amp;quot;Time Series with Outliers&amp;quot;)&lt;br /&gt;
boxplot(dat, main=&amp;quot;Boxplot Identifying Outliers&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Identifying and removing outliers based on extreme boxplot bounds&lt;br /&gt;
outlier_bounds &amp;lt;- boxplot.stats(dat)$out&lt;br /&gt;
dat_clean &amp;lt;- dat[!(dat %in% outlier_bounds)]&lt;br /&gt;
&lt;br /&gt;
# Re-plot clean data&lt;br /&gt;
par(mfrow=c(1,1))&lt;br /&gt;
plot.ts(dat_clean, main=&amp;quot;Cleaned Time Series&amp;quot;, col=&amp;quot;blue&amp;quot;, lwd=2)&lt;br /&gt;
&amp;lt;/source&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==Applications==&lt;br /&gt;
&lt;br /&gt;
1. '''Epidemiology and Public Health:''' Time series models are essential for tracking disease outbreaks (e.g., COVID-19, influenza). SARIMA models help forecast case counts, accounting for weekly seasonal cycles (e.g., higher reporting on weekdays) and long-term trends, thereby informing hospital resource allocation.&lt;br /&gt;
2. '''Health Economics:''' Forecasting hospital readmission rates, healthcare costs, and resource utilization over time using ARIMA and Exponential Smoothing to optimize budget planning.&lt;br /&gt;
3. '''Clinical Monitoring:''' Analyzing ECG readings or continuous glucose monitoring data. Signal processing techniques combined with MA filtering help isolate true physiological signals from high-frequency noise.&lt;br /&gt;
4. '''Macroeconomics:''' As investigated by Nelson and Plosser (1982), understanding whether macroeconomic time series (like GDP or stock prices) are stationary around a deterministic trend or non-stationary stochastic processes fundamentally changes how we interpret economic shocks and formulate policy.&lt;br /&gt;
&lt;br /&gt;
==Software==&lt;br /&gt;
&lt;br /&gt;
Modern time series analysis in R relies on a few key packages:&lt;br /&gt;
* '''stats''' (Base R): Contains foundational functions like `ts()`, `acf()`, `pacf()`, `arima()`, `HoltWinters()`, and `stl()`.&lt;br /&gt;
* '''forecast''' (now transitioning to '''fable'''): Provides `auto.arima()`, `checkresiduals()`, `forecast()`, and modern tidy time series frameworks (`tsibble`).&lt;br /&gt;
* '''astsa''': Applied Statistical Time Series Analysis, highly recommended for its companion datasets and straightforward functions like `acf2()` (combines ACF and PACF in one plot).&lt;br /&gt;
* '''tseries''': Provides unit root tests like `adf.test()` to check for stationarity.&lt;br /&gt;
&lt;br /&gt;
==Problems==&lt;br /&gt;
&lt;br /&gt;
1) Consider a signal-plus-noise model of the general form &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is Gaussian white noise with &amp;lt;math&amp;gt;\sigma^{2}_{w} = 1&amp;lt;/math&amp;gt;. Simulate and plot &amp;lt;math&amp;gt;n=200&amp;lt;/math&amp;gt; observations from each of the following two models.&lt;br /&gt;
(a) &amp;lt;math&amp;gt;x_{t} = s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/20 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(b) &amp;lt;math&amp;gt;x_{t}=s_{t}+w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;s_{t}=\begin{cases} 0 &amp;amp; \text{ if } t=1,2,\dots,100 \\ 10\exp \left\{ -(t-100)/200 \right\} \cos\left( 2\pi t/4 \right) &amp;amp; \text{ if } t=101,102,\dots,200 \end{cases}&amp;lt;/math&amp;gt;&lt;br /&gt;
(c) Compare the signal modulators &amp;lt;math&amp;gt;(a) \exp \left\{-t /20 \right\}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;(b) \exp \left\{-t/200 \right\}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=1,2,\dots,100&amp;lt;/math&amp;gt;. What is the primary difference in the persistence of the signal?&lt;br /&gt;
&lt;br /&gt;
2) (a) Generate &amp;lt;math&amp;gt;n=100&amp;lt;/math&amp;gt; observations from the autoregression &amp;lt;math&amp;gt;x_{t}=-0.9 x_{t-2} + w_{t}&amp;lt;/math&amp;gt;, with &amp;lt;math&amp;gt;\sigma_w=1&amp;lt;/math&amp;gt;. Next, apply the moving average filter &amp;lt;math&amp;gt;v_{t}=\left(x_{t}+x_{t-1}+x_{t-2}+x_{t-3} \right)/4&amp;lt;/math&amp;gt; to the data you generated. Now, plot &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; as a line and superimpose &amp;lt;math&amp;gt;v_{t}&amp;lt;/math&amp;gt; as a dashed line. Comment on the behavior of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and how applying the moving average filter changes that behavior. [Hint: use &amp;lt;code&amp;gt;v &amp;lt;- filter(x, rep(1/4,4), sides=1)&amp;lt;/code&amp;gt; for the filter].&lt;br /&gt;
(b) Repeat (a) but with &amp;lt;math&amp;gt;x_{t}=\cos \left( 2 \pi t/4 \right)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Repeat (b) but with added &amp;lt;math&amp;gt;N(0,1)&amp;lt;/math&amp;gt; noise, &amp;lt;math&amp;gt;x_{t}=\cos \left(2 \pi t/4 \right) + w_{t}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(d) Compare and contrast the smoothing effects observed in &amp;lt;math&amp;gt;(a) - (c)&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
3) For the two series, &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; in Problem 1 (a) and (b):&lt;br /&gt;
(a) Compute and plot the mean functions &amp;lt;math&amp;gt;\mu_{x}(t)&amp;lt;/math&amp;gt; for &amp;lt;math&amp;gt;t=1,2,\dots,200&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Explain why these models are not stationary.&lt;br /&gt;
&lt;br /&gt;
4) Consider the time series &amp;lt;math&amp;gt;x_{t} = \beta_{1} + \beta_{2} t + w_{t}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;\beta_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;\beta_{2}&amp;lt;/math&amp;gt; are known constants and &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt; is a white noise process with variance &amp;lt;math&amp;gt;\sigma^{2}_{w}&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Determine whether &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is stationary. Justify your answer mathematically.&lt;br /&gt;
(b) Show that the process &amp;lt;math&amp;gt;y_{t}=x_{t} - x_{t-1}&amp;lt;/math&amp;gt; is stationary.&lt;br /&gt;
(c) Show that the mean of the moving average &amp;lt;math&amp;gt;v_{t}= \frac{1}{2q+1} \sum_{j=-q}^{q} x_{t-j}&amp;lt;/math&amp;gt; is &amp;lt;math&amp;gt;\beta_{1} + \beta_{2} t&amp;lt;/math&amp;gt;, and provide a simplified expression for its autocovariance function.&lt;br /&gt;
&lt;br /&gt;
5) A time series with a periodic component can be constructed from &amp;lt;math&amp;gt;x_{t} = U_{1} \sin(2 \pi \omega_{0} t) + U_{2} \cos(2 \pi \omega_{0} t)&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;U_{1}&amp;lt;/math&amp;gt; and &amp;lt;math&amp;gt;U_{2}&amp;lt;/math&amp;gt; are independent random variables with zero means and &amp;lt;math&amp;gt;E(U^{2}_{1}) = E(U^{2}_{2}) = \sigma^{2}&amp;lt;/math&amp;gt;. The constant &amp;lt;math&amp;gt;\omega_{0}&amp;lt;/math&amp;gt; determines the period. Show that this series is weakly stationary with autocovariance function &amp;lt;math&amp;gt;\gamma(h) = \sigma^{2} \cos(2 \pi \omega_{0} h)&amp;lt;/math&amp;gt;. [Hint: you will need to refer to standard trigonometric identities for products of sine and cosine functions].&lt;br /&gt;
&lt;br /&gt;
6) Suppose we would like to predict a single stationary series &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; with zero mean and autocorrelation function &amp;lt;math&amp;gt;\gamma(h)&amp;lt;/math&amp;gt; at some time in the future, say &amp;lt;math&amp;gt;t+l&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;l&amp;gt;0&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) If we predict using only &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; and some scale multiple &amp;lt;math&amp;gt;A&amp;lt;/math&amp;gt;, show that the mean-square prediction error &amp;lt;math&amp;gt;MSE(A)=E \left[ (x_{t+l}-A x_{t})^{2} \right]&amp;lt;/math&amp;gt; is minimized by the value &amp;lt;math&amp;gt;A = \rho(l)&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Show that the minimum mean-square prediction error is &amp;lt;math&amp;gt;MSE(A)= \gamma(0) \left[ 1- \rho^{2}(l) \right]&amp;lt;/math&amp;gt;.&lt;br /&gt;
(c) Show that if &amp;lt;math&amp;gt;x_{t+l} = Ax_{t}&amp;lt;/math&amp;gt; perfectly, then &amp;lt;math&amp;gt;\rho(l) = 1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;gt; 0&amp;lt;/math&amp;gt;, and &amp;lt;math&amp;gt;\rho(l) = -1&amp;lt;/math&amp;gt; if &amp;lt;math&amp;gt;A &amp;lt; 0&amp;lt;/math&amp;gt;.&lt;br /&gt;
&lt;br /&gt;
7) Let &amp;lt;math&amp;gt;w_{t}&amp;lt;/math&amp;gt;, for &amp;lt;math&amp;gt;t=0, \pm 1, \pm 2, \dots&amp;lt;/math&amp;gt; be a normal white noise process, and consider the series &amp;lt;math&amp;gt;x_{t}=w_{t} w_{t-1}&amp;lt;/math&amp;gt;. Determine the mean and autocovariance function of &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt;, and state whether it is strictly stationary, weakly stationary, or neither.&lt;br /&gt;
&lt;br /&gt;
8) Suppose &amp;lt;math&amp;gt;x_{1}, x_{2}, \dots, x_{n}&amp;lt;/math&amp;gt; is a sample from the process &amp;lt;math&amp;gt;x_{t} = \mu + w_{t} - 0.8 w_{t-1}&amp;lt;/math&amp;gt;, where &amp;lt;math&amp;gt;w_{t} \sim \text{WN}(0, \sigma^{2}_{w})&amp;lt;/math&amp;gt;.&lt;br /&gt;
(a) Show that the mean function is &amp;lt;math&amp;gt;E(x_{t})= \mu&amp;lt;/math&amp;gt;.&lt;br /&gt;
(b) Calculate the standard error of &amp;lt;math&amp;gt;\bar{x}&amp;lt;/math&amp;gt; for estimating &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt;. &lt;br /&gt;
(c) Compare your result in (b) to the case where &amp;lt;math&amp;gt;x_{t}&amp;lt;/math&amp;gt; is pure white noise. Is the standard error smaller or larger? Explain the result in the context of positive/negative autocorrelation.&lt;br /&gt;
&lt;br /&gt;
9) Although the model in Problem 1 (a) is not stationary, the sample ACF can still be computed and informative. For the data you generated in that problem, calculate and plot the sample ACF. Why does the ACF fail to decay to zero rapidly?&lt;br /&gt;
&lt;br /&gt;
10) (a) Simulate a series of &amp;lt;math&amp;gt;n=500&amp;lt;/math&amp;gt; Gaussian white noise observations and compute the sample ACF, &amp;lt;math&amp;gt;\hat{\rho}(h)&amp;lt;/math&amp;gt;, to lag 20. Compare the sample ACF you obtain to the theoretical ACF &amp;lt;math&amp;gt;\rho(h)&amp;lt;/math&amp;gt;. Do approximately 95% of the sample autocorrelations fall inside the bounds &amp;lt;math&amp;gt;\pm 2/\sqrt{500}&amp;lt;/math&amp;gt;?&lt;br /&gt;
(b) Repeat part (a) using only &amp;lt;math&amp;gt;n=50&amp;lt;/math&amp;gt;. How does changing &amp;lt;math&amp;gt;n&amp;lt;/math&amp;gt; affect the variability of the sample ACF and the reliability of the confidence bounds?&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
* Cryer, J. D., &amp;amp; Chan, K. S. (2008). ''Time Series Analysis: With Applications in R''. Springer.&lt;br /&gt;
* Hyndman, R. J., &amp;amp; Athanasopoulos, G. (2021). ''Forecasting: Principles and Practice'' (3rd ed.). OTexts. (Available freely online at otexts.com/fpp3)&lt;br /&gt;
* Nelson, C. R., &amp;amp; Plosser, C. I. (1982). Trends and random walks in macroeconomic time series: Some evidence and implications. ''Journal of Monetary Economics'', 10(2), 139-162.&lt;br /&gt;
* Shumway, R. H., &amp;amp; Stoffer, D. S. (2017). ''Time Series Analysis and Its Applications: With R Examples'' (4th ed.). Springer.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php?title=SMHS_TimeSeries}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18362</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18362"/>
		<updated>2026-03-30T00:12:21Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2026 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [https://wiki.socr.umich.edu/images/6/69/Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2023: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=File:Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf&amp;diff=18361</id>
		<title>File:Dinov AI OpenScience DataSharing RCR DCMB Bioinfo504 2026.pdf</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=File:Dinov_AI_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_2026.pdf&amp;diff=18361"/>
		<updated>2026-03-30T00:12:01Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18360</id>
		<title>SOCR News</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News&amp;diff=18360"/>
		<updated>2026-03-30T00:11:45Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* 2026 */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_NewsEventsAnnouncements | SOCR News, Events &amp;amp; Announcements]] - News &amp;amp; [http://www.socr.umich.edu/html/feed.rss RSS Feeds]==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==2026==&lt;br /&gt;
* June 1-3, 2026: [https://socr.umich.edu SOCR] is ''co-organizing'' and ''co-sponsoring'' the [https://www.statsinimaging.org/SMI-2026/ 2026 ASA-SMI Conference]. [https://www.statsinimaging.org/SMI-2026/ Jian Kang, Ivo Dinov, and Tim Johnson are chairing the 2026 ASA Statistical Methods in Imaging (SMI) Symposium, at the University of Michigan, Ann Arbor, MI].&lt;br /&gt;
&lt;br /&gt;
* March 30, 2026: Ivo Dinov presented [ Data Sharing &amp;amp; Open, Rigorous, Reproducible Science], BIOINF-504: Rigor and Transparency to Enhance Reproducibility, University of Michigan, 3813/17 Med Sci II Bldg.&lt;br /&gt;
&lt;br /&gt;
* March 23, 2026: Ivo DInov is presenting [[SOCR_FAMU_Talk_2026 | '''The Realities of Augmented Intelligence''']] at the [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications semiar series]. &lt;br /&gt;
&lt;br /&gt;
* March 15-20, 2026, [https://summit.aps.org/attend/abstracts/ APS Global Physics Summit, Denver, CO], [[SOCR_News_APS_GPS_2026 | details on SOCR Spacekime/TCIU short course, lectures and sessions at APS GPS details are here]].&lt;br /&gt;
&lt;br /&gt;
* January 2026: [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] is elected as the [https://engage.aps.org/gds/home 2026-2027 Vice-Chair of American Physical Society (APS) Group on Data Science (GDS)].&lt;br /&gt;
&lt;br /&gt;
* January 8, 2026: SOCR-MDP Jumpstart Kickoff Event. [https://forms.gle/VgDJRmKJQdSi8ftP9  RSVP to participate]. Logistics: January 8th, 6:00pm – 8:30pm, [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-18 NCRC, North Campus Research Complex, Building 18]. &lt;br /&gt;
: Agenda:&lt;br /&gt;
:: 6:00pm Jumpstart Registration, Check-in, Grab Dinner, Mingle with your team&lt;br /&gt;
:: 6:15pm Dinner + Welcome &amp;amp; Presentation by Ellen Solomon, Faculty Research Program Manager&lt;br /&gt;
:: 6:30-8:30pm Team Breakouts: Individual Team Discussions&lt;br /&gt;
:: SOCR Team introductions&lt;br /&gt;
:: Review team logistics (meeting date, time, location)&lt;br /&gt;
:: Review team expectations&lt;br /&gt;
:: Take a team photo!&lt;br /&gt;
&lt;br /&gt;
* January 4-7, 2023: At the annual [https://jointmathematicsmeetings.org/jmm 2026 JMM Congress], Ivo Dinov is presenting a talk, in [[SOCR_JMM_2026 | AMS/JMM Special Session (Session ID: 1675) on ''AMS Special Session on Mathematical Foundation of Machine Learning'']], organized by Maryam Bagherian &amp;amp; Emanuele Zappala (Idaho State University).&lt;br /&gt;
&lt;br /&gt;
==2025==&lt;br /&gt;
&lt;br /&gt;
* [https://clnq.blogspot.com/ SOCR CLNQ-blogger Blog] and [https://tmd-ai.blogspot.com 360TMD-AI Blog].&lt;br /&gt;
[[Image:MonaLisa_Orig_v_AI.png|250px|thumbnail|right| GAIM Gen-1: ''Reproducing and improving existing human art'' (Mona Lisa)]]&lt;br /&gt;
[[Image:GAIM_Synth_MotherMonaLisa_AI.png|250px|thumbnail|right| GAIM Gen-2: ''De novo content creation'' (Mother Mona Lisa)]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png|250px|thumbnail|right| (V.1) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150_V2.png|250px|thumbnail|right| (V.2) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/GAIM/ SOCR GAIM], ca. ~2150]]&lt;br /&gt;
[[Image:GAIM_ChronoSculptureAnime3_xs.gif|thumbnail|right| (V.3) GAIM Gen-3: ''Autonomous Creative Science and Arts'' [https://socr.umich.edu/docs/uploads/2025/GAIM_ChronoSculptureAnime3.gif Chromo-Sculpture Anime], ca. ~2150]]&lt;br /&gt;
* September 2025: SOCR ''example visually demonstrating the abilities of contemporary'' [https://socr.umich.edu/GAIM/ ''Generative AI Models (GAIM)/Augmented Intelligence Models (AIA)''] with progressively multi-generational improvements:&lt;br /&gt;
** Gen-1: ''Reproducing and improving existing human art'': Identify the [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated''] [https://en.wikipedia.org/wiki/Mona_Lisa Mona Lisa] paintings. &lt;br /&gt;
** Gen-2: ''De novo content creation'': [https://wiki.socr.umich.edu/images/4/4a/AI_SynthDaVinci_MonaLisa2.pdf completely AI-generated Renaissance atelier scene], which depicts a noble lady (mother) with an expressive and directive gesture, actively overseeing the commissioned painting of her daughter in the background by an artist. In both AI-generated scenes, the lady's face and posture reflect her authority and noble status, while her daughter is shown seated, calmly posing for the portrait, with her presence reinforcing the mother’s directive role. A Leonardo DaVinci-like painter is at work with an easel, palette, and sketches around him, inside a warm and atmospheric Renaissance atelier with wooden beams and muted light. The entire composition uses warm Renaissance tones, soft sfumato shading, and layered detail to capture the cultural essence of the 15th–16th century while emphasizing action and interaction, not just passive posing like in [https://wiki.socr.umich.edu/images/5/50/MonaLisa_Original_v_AI.pdf ''real'' (by Leonardo Da Vinci) vs. the ''AI-generated'' scene].&lt;br /&gt;
** Gen-3: ''Autonomous Creative Science and Arts'': GAIMs/AIAa can generate realistic autonomously creative art of new prospective, ubiquitous form of human visual art in the year 2150. [https://wiki.socr.umich.edu/index.php/File:AI_ChronoSculpting_TemporalKinetic_Augmentation_SynthArt_2150.png This image] shows realistic (high-precision and detailed) painting of a futuristic ''Chrono-Sculpture'', a form of AI-predicted future 4D work of art anchored to a specific physical location, composed of light, data, and algorithmically generated form, which evolves over time based on a complex set of inputs. This is an AI-speculative forecast of a future Primary Art Form, Chrono-Sculpting (Temporal Kinetic Augmentation). &lt;br /&gt;
:: By 2150, the distinction between digital and physical reality may be expected to become functionally meaningless for most of the population, who experience the world through ambient, non-invasive neural or optical interfaces. This mixed-reality environment is the canvas for a new form of a dominant visual art form Chrono-Sculpting, which does not have a physical composition. Its substance is a localized, computationally intensive projection of ''phased photons'' and ''contextual data streams'', rendered in real-time for each viewer. &lt;br /&gt;
:: Visually, a Chrono-Sculpture appears as a polychromatic, semi-translucent, and kinetically fluid form. Its shape is never entirely static, often resembling a slow-motion capture of smoke, a living fractal, or a liquid crystal formation. Its texture can range from smooth and ethereal to sharp and crystalline, and its opacity might shift, allowing the physical world behind it to be an integral part of the composition. The experience is multi-sensory, often accompanied by localized spatial audio or even subtle haptic feedback for those with compatible interfaces.&lt;br /&gt;
:: The color palette is not fixed but is a dynamic function of input data. For example, a sculpture in a public park might translate the local atmospheric pressure, pollen count, and ambient noise level into a constantly shifting color field. An artist might map the emotional sentiment of local network traffic to a spectrum from deep violets (negative) to brilliant golds (positive). The &amp;quot;value&amp;quot; in the artistic sense is derived from the harmony and poignancy of these data-driven transformations.&lt;br /&gt;
:: The ''shape'' of a Chrono-Sculpture is a ''probability cloud of form''. The artist does not define a single, static shape. Instead, they design a set of rules and a constrained set of aesthetic possibilities within which the sculpture exists and evolves.&lt;br /&gt;
:: ''Ontological Kernel'': The core of the sculpture is its ontological kernel - the foundational algorithm and aesthetic constraints designed by the human artist. This is the artist's unique signature and vision. It defines the sculpture's behavioral tendencies, its potential geometric vocabulary (e.g., biomorphic curves, sharp crystalline structures), and its response logic to data inputs.&lt;br /&gt;
:: ''Temporal Dimension'': This is the crucial fourth dimension. A sculpture might be programmed to ''grow over time'', e.g., reflecting the changing seasons. It might relive a key historical event associated with its location every day at noon, with the data from each new day subtly altering the re-enactment. The artwork is a performance unfolding over decades.&lt;br /&gt;
:: The creation of a Chrono-Sculpture is a process of ''systems architecture and aesthetic curation'', not direct manipulation. The artist acts more like a choreographer or a gardener than a traditional painter or sculptor.&lt;br /&gt;
:: ''Site Anchoring &amp;amp; Historical Curation'': The artist first selects a physical location (the ''anchor''). They then curate vast datasets relevant to that site—geological surveys, historical archives, demographic shifts, ecological data, etc. This is a deeply humanistic and research-intensive phase.&lt;br /&gt;
:: ''Algorithmic Weaving'': Using highly advanced AI as a collaborative tool (often referred to as a latent space brush), the artist designs the ontological kernel. They do not code in a traditional sense but rather guide the AI through aesthetic choices, defining the relationships between data inputs and visual/auditory outputs. The goal is to imbue the system with a specific character or ''soul''. The process can be described by a simplified function: \(A(t) = f_{\theta}(L, D(t), V(t))\), where \(A(t)\) is the state of the artwork at time \(t\); \(f_{\theta}\) is the kernel function designed by the artist with parameters \(\theta\); \(L\) is the static, curated data set for the location; \(D(t)\) is the set of real-time dynamic data streams (e.g., weather, network traffic); and \(V(t)\) is the data from viewers interacting with the piece (e.g., gaze duration, proximity, biometric feedback).&lt;br /&gt;
:: ''Neuro-Kinetic Tuning'': The artist rehearses with the sculpture in a simulated environment, often using a direct brain-computer interface (BCI). They experience the sculpture's evolution and refine its responses, not by rewriting code, but by providing direct neural feedback of approval or dissonance, tuning the parameters \(\theta\) until the sculpture's behavior aligns with their artistic intent.The finalized kernel is uploaded to the global mixed-reality mesh and permanently anchored to its physical coordinates, where it begins its long, evolving existence.&lt;br /&gt;
:: ''Value and Appeal'': The ubiquity and appreciation for Chrono-Sculpting stem from several key factors that address a post-industrial, data-saturated society's needs: (1) Re-enchantment of Place, in a globally connected world, Chrono-Sculpting re-invests physical locations with deep, unique, and evolving meaning. A simple street corner can become a profound historical and aesthetic experience; (2) Living Art, it is fundamentally anti-static. The desire to see &amp;quot;what the sculpture is doing today&amp;quot; drives repeat engagement. Communities form around observing and interpreting the long-term behavior of their local public sculptures; (3) Authenticity of Intent, while AI is a core part of the toolset, the value is placed entirely on the human artist's vision. The genius lies not in crafting an object, but in designing a beautiful, meaning-making system. The ontological kernel is the revered artifact, and its elegance and depth are what critics assess; and (4) Participatory Experience, viewers subtly influence the art through their presence and attention, creating a gentle, subconscious dialogue between the artist, the art, the public, and the place itself. This resolves the modernist tension between the artwork and the viewer, making the viewer a part of the artwork's environment. This art form represents a synthesis of land art, generative art, performance art, and data visualization. It is an art not of static objects, but of dynamic, living systems, reflecting a future where the boundary between information and reality has beautifully and irrevocably blurred.&lt;br /&gt;
&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting an invited talk at the [[SOCR_News_HDDA14_2025 | ''14th High Dimensional Data Analysis Workshop (HDDA-XIV); CMU Biological Station on Beaver Island, Lake Michigan'']].&lt;br /&gt;
* August 19-22, 2025: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/1c/Dinov_Spacekime_2025_Slidedeck_HDDA_2025.pdf Spacekime Representation, Statistical Inference &amp;amp; AI prediction using Repeated Measurement Longitudinal Data] at the [https://www.cmich.edu/academics/colleges/college-science-engineering/departments-schools/statistics-actuarial-and-data-sciences/HDDA 2025 High-Dimensional Data Analysis Conference (HDDA-14)], hosted at the Central Michigan University Biological Station on Beaver Island.&lt;br /&gt;
* August 20, 2025: Ivo Dinov presented [https://wiki.socr.umich.edu/images/c/cf/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2025.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan, [https://maps.studentlife.umich.edu/building/central-campus-classroom-building CCCB #0420].&lt;br /&gt;
* August 2-7, 2025: Ivo Dinov is organizing an [[SOCR_News_ISS_JSM_2025 | Invited Special Session (#0127), ''Statistical Inference and AI Modeling of High-Dimensional Longitudinal Data'', at the 2025 Joint Statistics Meeting (JSM) in Nashville, TN]].&lt;br /&gt;
* June 4, 2025: [https://nursing.umich.edu/about/news-portal/202506-umsn-building-future-ai-health-care UMSN News article ''The Future of AI in Health Care''] discusses recent [https://socr.umich.edu/HTML5/ SOCR AI developments], such as [https://ilp.statisticalcomputing.org/ ILP], [https://mita.statisticalcomputing.org/ MITA], and [https://NAIT.statisticalcomputing.org/ NAIT].&lt;br /&gt;
* May 15, 2025: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] Oral Doctoral Dissertation Defense [https://wiki.socr.umich.edu/images/9/97/Shen_PhD_Defense_May15_2025_Flier.pdf Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics], Thursday, May 15, 2025, 9 AM ET, [https://my.matterport.com/show/?m=tQjegaZqoHf 3D View of Taubman Library 2nd floor], [https://maps.studentlife.umich.edu/building/alfred-a-taubman-health-sciences-library 2903 Taubman Health Sciences Library (THSL), 2nd Floor]. &lt;br /&gt;
** Update: [https://www.socr.umich.edu/people/ SOCR] Welcomes [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html '''Dr. Yueyang Shen'''] to the [https://www.socr.umich.edu/people/dinov/bio.html Sina-Bernoulli-Euler-Laplace-Huffer-Sumners Mathematics Genealogy Linage], see this partial dynamic graph illustrating [https://socr.umich.edu/docs/uploads/Dinov_MathGeneologyTree_2006.pdf Prof. Dinov's academic heritage].&lt;br /&gt;
** [https://forms.gle/fjbBTPK8JQoQ69qXA RSVP by Saturday 5/17/25] to join us in person to congratulate Dr. Yueyang Shen on the completion of his Doctorate Degree. The graduation celebration will include a luncheon Thu 5/22/25, 12:30 PM, [https://maps.studentlife.umich.edu/building/school-of-nursing-building-2 SNB2].&lt;br /&gt;
** Dr. Shen's dissertation, [https://dx.doi.org/10.7302/26331 ''Complex Time Representation and Observability of Repeated Measurement Processes with Applications to Spacekime Analytics'', is available here].&lt;br /&gt;
* May 7, 2025: Ivo Dinov is presenting a talk on [https://wiki.socr.umich.edu/images/1/15/Dinov_BSI_Lecture_BrainAI_May_2025_Slidedeck.pdf ''Brain Technologies – From Quanta to Neural Nets – Multiscale AI Systems using Quantumscale, Nanoscale, Microscale, and Macroscale Brain Networks Data''] at the 2025 Biosciences Institute Challenge, ''Nanoscale, Microscale, and Macroscale Networks in Brain Health Technologies'', here is [https://socr.umich.edu/docs/uploads/2025/ID_MultiscaleBrainTechnologies_FromQuantumToAI.mp3 an AI-generated podcast of the lecture (11-min audio/mp3)].&lt;br /&gt;
* May 01, 2025: Nominated by the [https://www.amstat.org/ AmStats], [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] is appointed to serve a term on the [https://www.nature.com/nm/statistics-advisory-panel Nature Medicine Statistical Advisory Board].&lt;br /&gt;
* April 25, 2025: Ivo Dinov is presenting at the [[SOCR_APS_GDS_VTS_SK_2025 | ''APS GDS Virtual Tutorial Series'' on Complex-time (kime) Representation, Statistical Inference, and AI prediction]].&lt;br /&gt;
* March 15-21, [https://summit.aps.org/ 2025 APS Summit], Anaheim, CA: Yueyang Shen, Yupeng Zhang, and Ivo D. Dinov are presenting on [[SOCR_News_APS_2025 |''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' and ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'']].&lt;br /&gt;
* January 9, 2025: [[SOCR_MDP_Jan_2025_JumpStart | SOCR-MDP 2025 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2024==&lt;br /&gt;
* September 1, 2024: Ivo Dinov is [https://regents.umich.edu/files/meetings/07-24/2024-07-IV-1.pdf appointed by the UMich Board of Regents as the Henry P. Tappan Collegiate Professor] (September 1, 2024 through August 31, 2029). The [https://bentley.umich.edu/legacy-support/um/umpresid.php second UMich President], [https://en.wikipedia.org/wiki/Henry_Philip_Tappan Dr. Henry P. Tappan] (1852-1863), was a [https://heritage.umich.edu/stories/tappans-end/ prolific writer who provided a vision for establishing the University of Michigan as a primer research intensive academic institution].&lt;br /&gt;
* August 28-30, 2024: Ivo Dinov is organizing a special session [[SOCR_News_HDDA_2024| Data Science, Artificial Intelligence, and High-Dimensional Spatiotemporal Dynamics]] at the 2024 High-Dimensional Data Analysis Conference (HDDA-13) in Singapore.&lt;br /&gt;
* July 8-10, 2024: (2024 NIGMS TWD PDO Biomedical Training Program Directors' Conference) [https://sites.udel.edu/grimesgroup/research-team/ Catherine Grimes] (Delaware) and [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (Michigan) are co-Chairing a [https://events.faseb.org/event/dc343624-9195-4916-96f8-234b8c15b6f0/ Special Session of the T32 Predoctoral Program PDs/PIs].&lt;br /&gt;
** A SOCR Analytic Report of the [https://forms.gle/F6dQLqc6f1HV73NT8 pre-survey of NIHMS T32 Predoctoral Training Program PIs] is available in [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.html HTML] and [https://socr.umich.edu/docs/uploads/2024/SOCR_NIHMS_TWD_Predoc_T32_PI_Survey_2024_Analytics.pdf PDF] formats.&lt;br /&gt;
** [https://docs.google.com/presentation/d/1p9qfcJOSU_HN3fA7iJlBUGmutVv4zhtt/?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Session GSlidedeck]&lt;br /&gt;
* May 29-31, 2024: (ASA SMI Annual Meeting) Yueyang Shen is giving a talk on [[SOCR_News_AmStats_SIM_2024 |''Statistical Foundations of Invariance in Deep Network Learning'' ('''2024 Best Paper Award!''')]]. Ivo Dinov is presenting a short course tutorial on ''Spacekime Analytics'' and organizing a Special Session on [[SOCR_News_AmStats_SIM_2024 |  ''Longitudinal Imaging and Biostatistical Methods'', at the 2024 American Statistical Association's Statistics in Imaging Annual Meeting]] in Indianapolis, IN.&lt;br /&gt;
** [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] and [https://math.wisc.edu/graduate-students/ Yupeng Zhang] receive the [https://www.statsinimaging.org/smi_competitions_all/ 2024 annual '''&amp;quot;Theory and Methods Track&amp;quot; Best Student Paper Award'''] at the [https://medicine.iu.edu/biostatistics/news-events/statistical-methods-in-imaging-conference 2024 American Statistical Association (AmStat) - Statistical Methods in Imaging (SMI) Conference]. &lt;br /&gt;
:: This paper [https://socr.umich.edu/docs/uploads/2024/Shen_StatisticalFoundationsOfInvarianceInDNN_SMI_2024.pdf ''Statistical foundations of invariance and equivariance in deep artificial neural network learning''] was recognized as being of '''remarkable quality''' by the AmStat/SMI review committee.&lt;br /&gt;
* May 29–August 9, 2024: Announcement of the [https://globalhealthequity.umich.edu/graduate-research-internship-projects#novel-ai-techniques-for-exploring-health-equity ''SOCR Graduate Student Summer Research Internship and Training Program in Global Health Equity'']. This program is open to UMich graduate students in any discipline, including master’s, doctoral, and other professional degree programs. This program is based at the U-M Ann Arbor campus. Students will need to be available 20 hours per week from May 29–August 9, 2024. [https://globalhealthequity.umich.edu/graduate-student-summer-research-internship-training-program Application Deadline is February 16, 2024].&lt;br /&gt;
* May 6, 2024: [https://drive.google.com/drive/folders/1GC-IPAfNR-42mecWZq_-J0VuW3M29n8z BIDS-TP Hackathon tackling a pair of biomedical informatics, data science and artificial intelligence challenges].&lt;br /&gt;
* April 25, 2024: Ivo Dinov is giving a lecture at the [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?catId=90 2024 OLLI Lecture Series on Artificial Intelligence] (AI), [https://mmcommunityprograms.med.umich.edu/umich/course/course.aspx?C=374&amp;amp;pc=9&amp;amp;mc=90&amp;amp;sc=0 AI in Health – Research Promises, Education Perils and Clinical Practice Impact], [https://www.wccnet.edu/visit/room-locator/ml.php Towsley Auditorium], [https://g.page/WashtenawCC?share Washtenaw Community College], [https://wiki.socr.umich.edu/images/2/21/Dinov_OLLI_Lecture_AI_in_Health_April25_2024_Slidedeck.pdf PDF Slidedeck].&lt;br /&gt;
* April 15, 2024: [[MIDAS_GDSC_Program_2024_Graduation | MIDAS GDSC Program - 2024 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 4, 2024: Ivo Dinov is presenting [[SOCR_News_APS_GDS_April_2024 | ''Complex-time Representation of Repeated Measurement Longitudinal Data and Space-kime Analytics'' at the April 2024 APS Annual Meeting in Sacramento, CA]].&lt;br /&gt;
* March 3-8, 2024: Ivo Dinov is organizing an [[SOCR_News_APS_GDS_2024 | APS Special Invited Session on ''Model-based Statistical Physics, Computable Data, and Model-Free Artificial Intelligence'' at the March 2024 APS Annual Meeting in Minneapolis, MN]].&lt;br /&gt;
* January 19–20, 2024, Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/f/f3/Dinov_BrainCancer_Lecture_AI_Jan_2024_Slidedeck.pdf ''AI Bio-Innovations in Health &amp;amp; Neurooncology - Research Promises, Learning Opportunities &amp;amp; Clinical Practice Impact''] at the [https://bibraincancer.umich.edu/4340-2/ 2024 Bioengineering in Brain Cancer Challenge], at the [https://bibraincancer.umich.edu/ University of Michigan Biointerfaces Institute], UM North Campus Research Complex (NCRC).&lt;br /&gt;
* January 11, 2024: [[SOCR_MDP_Jan_2024_JumpStart SOCR-MDP|2024 JumpStart Event]], Time: 6:00-8:30 pm, dinner will be provided, SOCR Team meets at NCRC Building – NCRC10 Room #G064.&lt;br /&gt;
&lt;br /&gt;
==2023==&lt;br /&gt;
* December 11, 2023. Ivo Dinov was interviewed for [https://www.techtarget.com TechTarget] on [https://www.techtarget.com/searchenterpriseai/feature/How-generative-AI-could-change-healthcare ''How generative AI could change healthcare''], by [https://www.techtarget.com/contributor/Mary-K-Pratt Mary K. Pratt], an award-winning freelance journalist.&lt;br /&gt;
* December 1, 2023. Ivo Dinov is moderating a [[SOCR_Events_Ross_GAIM_in_Healthcare_2023 | Generative AI in Healthcare Panel at the University of Michigan Ross School of Business, Business+Tech Training Program]], 11:45 AM ET.&lt;br /&gt;
* October 24, 2023. Ivo Dinov presented the [https://wiki.socr.umich.edu/images/a/a7/Dinov_DCMB_BGP_PhD_DoctoralStudents_2023_Outline.pdf Statistics Online Computational Resource at the 2023 DCMB Bioinformatics Graduate Program], 11:30 AM ET, at [https://maps.studentlife.umich.edu/building/medical-science-unit-ii 3813/3817 Med Sci II, UMich/MM]. &lt;br /&gt;
* October 21, 2023: Yueyang Shen presented [https://wiki.socr.umich.edu/images/b/b6/Shen_APS_EGLS_SufficiencyExchangeability_InvarianceTalk_2023.pdf ''Invariance and equivariance in deep network learning: mathematical representation, probabilistic symmetry, variable exchangeability &amp;amp; sufficient statistics''] at the Fall 2023 Meeting of the [https://artsandsciences.csuohio.edu/physics/aps-eastern-great-lakes-section-egls APS Eastern Great Lakes Section (EGLS), October 20-21, 2023, Cleveland State University, Cleveland, Ohio].&lt;br /&gt;
* October 16-17, 2023: Ivo Dinov is presenting a talk on ''AI in Biomedicine &amp;amp; Health – Research Promises, Education Perils &amp;amp; Clinical Practice Impact'' at the [[SOCR_Events_MWACD_BRCF_2023 | 2023 Annual Meeting of the Midwest Association of Core Directors (MWACD) is a chapter of the international Association of Biomolecular Resource Facilities (ABRF)]].&lt;br /&gt;
* Sept 1, 2023: [[SOCR_MDP_Sept_2023_Retreat | SOCR-MDP 2023 Fall Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* August 22, 2023: [https://drive.google.com/drive/folders/1DYU156qwPsGG9-0MDSktqy5fHA-fWFgP BIDS-TP Hackathon] tackling a pair of biomedical informatics, data science and artificial intelligence challenges.&lt;br /&gt;
* August 23, 2023: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/07/Dinov_OpenScience_DataSharing_RCR_DCMB_Bioinfo504_Aug_2023.pdf ''Data Sharing &amp;amp; Open, Rigorous, Reproducible Science''] at the [https://medicine.umich.edu/dept/dcmb/education/course-descriptions BIOINF-504: Rigor and Transparency to Enhance Reproducibility], University of Michigan.&lt;br /&gt;
* July 01, 2023: Read the [https://spacekime.wordpress.com/2023/07/01/spacekime-explained-by-generative-ai/ first generative artificial intelligence model (GAIM) interpretaiton of &amp;quot;spacekime analytics&amp;quot;] using OpenAI and Google PaLM models. It includes textual descriptions of spacekime, as well as, pictorial [https://spacekime.files.wordpress.com/2023/07/4dspacekimeuniverse_gaim_generated_p1_2023.png GAIM-generated image renditions].&lt;br /&gt;
* June 26-28, 2023: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/8/86/Dinov_APS_DSECOP_UMD_June2023.pdf Data Science Modules Enhancing the Biophysics Curriculum] at the [https://engage.aps.org/gds/home APS Group of Data Science] [https://dsecop.org/workshops/ 2023 DSECOP Workshop at the University of Maryland], see [https://dsecop.org/assets/23_Workshop/DSECOP23_Schedule.pdf the agenda].&lt;br /&gt;
* May 15, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/14gsIxDJ6KbYdazlOGkHdFe5dBkQe1Ksg/edit?usp=sharing&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true ''Faculty Development - Tracking Research and Scientific Productivity''].&lt;br /&gt;
* April 26, 2023: Yueyang Shen is presenting [[SOCR_APS_Apr_2023_ILT_Shen | ''Numerical Methods and Analysis for Computing Forward and Inverse Laplace Transform For discrete and continuous signals'', at the April 2023 Meeting of the American Physical Society (APS)]].&lt;br /&gt;
* April 19, 2023: [[SOCR_MDP_Apr_2023_Retreat | SOCR-MDP 2023 April Retreat]], Time: 5:00-6:30 PM, dinner will be provided. All [https://www.socr.umich.edu/people/ SOCR Team members are invited to attend].&lt;br /&gt;
* April 14, 2023: [[MIDAS_GDSC_Program_2023_Graduation|MIDAS GDSC Program - 2023 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 3-4, 2023: Ivo Dinov is presenting [https://docs.google.com/presentation/d/1HCCWtRaqtPVMLAAcTiO6ankF6yLdLu0NUHxhpSfHIDg/edit?usp=sharing Nursing and Healthcare 2030: AI Promises &amp;amp; Perils in Education, Scholarship &amp;amp; Clinical Practice] at the [https://nursing.umich.edu/research/research-day 2023 UMSN Research Day].&lt;br /&gt;
* March 14, 2023: Ivo Dinov is presenting [[SOCR_AmStat_SII_2023 | Data Science and Predictive Analytics (DSPA2)'']] at the [https://community.amstat.org/statisticsinimagingsection/home March Workshop of the Statistics in Imaging (SII) Section].&lt;br /&gt;
* March 15, 2023: [[SOCR_News_GenAI_RD_March_2023 |SOCR R&amp;amp;D in generative artificial intelligence (gen-AI)]].&lt;br /&gt;
* March 4, 2023: Prof. Dinov is selected as one of the [https://news.umich.edu/chatgpt-u-m-experts-can-discuss-ai-chatbots-their-reach-impact-concerns-potential/ University of Michigan Experts on generative artificial intelligence, gen-AI, (e.g., ChatGPT)]. &lt;br /&gt;
* March 1, 2023: Release of V.1.5 of the [https://rcompute.nursing.umich.edu/SOCR_AI_Bot/ SOCR AI Bot, which utilizes generative-AI for synthetic medical text generation, simulation of 2D medical images, and generation of code], [https://drive.google.com/file/d/1DtVA8wl-Tc8PYVzrOFcAz6AOLAve_WgA/view?usp=share_link video 1 (3-min)], [https://drive.google.com/file/d/1GS3ohrIgy0eJ1qfdTe7L2m0ShNUEmOSQ/view?usp=sharing video 2 (12-min)], [https://docs.google.com/presentation/d/1qzIL7acG8ugNoiTFb7Xjec4wc0BNmbKF/edit?usp=share_link&amp;amp;ouid=103614380037368892595&amp;amp;rtpof=true&amp;amp;sd=true Pressure Injury Slidedeck (2023)],  [https://drive.google.com/file/d/1FhdMXN9cvDD5NyMjZt8d2QzVgSTcnIPL/view?usp=sharing AI Bot Demo Script].&lt;br /&gt;
* January 5, 2023: [[SOCR_MDP_Jan_2023_JumpStart | SOCR-MDP 2023 JumpStart Event]], Time: 6:00 -8:30 pm], dinner will be provided, [https://www.socr.umich.edu/people/ SOCR Team] meets at [https://maps.studentlife.umich.edu/building/north-campus-research-complex-building-10 NCRC Building – NCRC10 Room #G064].&lt;br /&gt;
* January 4-7, 2023: Ivo Dinov and Joshua Welch are organizing a [[SOCR_JMM_2023 | AMS/JMM Special Session on ''Tensor Representation, Completion, Modeling and Analytics of Complex Data'']] at the [https://www.jointmathematicsmeetings.org/2270_intro 2023 JMM Congress], Boston, MA.&lt;br /&gt;
&lt;br /&gt;
==2022==&lt;br /&gt;
* December 17-19, 2022: Ivo Dinov, Yueyang Shen, and Milen V. Velev are presenting [[SOCR_News_CMStatistics2022 | ''Quantum Mechanics Uncertainty, Data Science Inference, and AI in Complex Time (Kime)'']] at the [http://www.cmstatistics.org/CMStatistics2022/index.php 15&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; International Computational and Methodological Statistics (CMStatistics) Conference], King's College London.&lt;br /&gt;
* October 25, 2022: Ivo Dinov presented [[SOCR_News_ASA_StatsInImaging_2022 | Complex-time (Kime) Representation of Longitudinal Data and Spacekime Analytics]] at the ASA Statistics in Imaging Section Seminar Series.&lt;br /&gt;
* October 22, 2022: Yueyang Shen is presenting [https://wiki.socr.umich.edu/images/5/58/YueyangShen_Laplace_Transform_APS_Talk_Oct_2022.pdf Laplace Transform and Inverse Laplace Transforms - Numerical Methods, Groups, and Clifford Algebra] at the [https://meetings.aps.org/Meeting/EGLSF22/Content/4321 Fall 2022 Meeting of the Eastern Great Lakes Section of the American Physical Society (EGLS of APS)], at [https://www.ltu.edu/arts_sciences/egls-schedule.asp Lawrence Tech University].&lt;br /&gt;
* August 24, 2022: Ivo Dinov presented [https://wiki.socr.umich.edu/images/3/32/Dinov_OpenScience_DataSharing_RCR_DCMB_Aug_2022.pdf Data Sharing &amp;amp; Open, Rigorous, Reproducible Science] at the 2022 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2022: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2022 |2022 MIDAS Data Science for Biomedical Scientists Bootcamp]], residential event ([https://maps.studentlife.umich.edu/building/weiser-hall UM Campus, Weiser Hall]). &lt;br /&gt;
* June 22-24, 2022 [[DSECOP_Workshop_Maryland_2022 | DSECOP June 2022 Workshop - &amp;quot;Data Science Education in the Physics Curriculum”]], hosted at the Edward St. John Center, University of Maryland.&lt;br /&gt;
* June 2-3, 2022: Ivo Dinov is presenting [https://wiki.socr.umich.edu/images/1/10/Dinov_Spacekime_DCMB_Retreat_June2022.pdf Quantum Physics Interface to Data Science, Artificial Intelligence &amp;amp; Spacekime Analytics] at the [https://sites.google.com/umich.edu/dcmb-retreat/home DCMB 10-th year anniversary retreat], Maumee Bay, Oregon, OH 4361.&lt;br /&gt;
* May 17, 2022: Ivo Dinov is presenting ''LAI in Data-Driven, Evidence-based, and Technology-Enhanced Health Science'' at the [[SoN_ LAI_4_Day_Intensive_May_2022| 2022 Leadership, Analytics and Innovation (LAI) Master’s Program 4-Day Intensive]].&lt;br /&gt;
* April 15, 2022: [[MIDAS_GDSC_Program_2022_Graduation|MIDAS GDSC Program - 2022 Graduation (MS/PhD) event]].&lt;br /&gt;
* April 12, 2022: Ivo Dinov is presenting &amp;quot;''Teaching Data Science with Technology''&amp;quot; at the [[LTU_OpenEd_Workshop_April_2022|Lawrence Technological University – OpenEd Workshop]].&lt;br /&gt;
* March 17, 2022: Ivo Dinov is presenting a [[SOCR_News_APS_Dinov_Spacekime_March_2022| ''Quantum Physics Interface to Data Science, Artificial Intelligence, and Spacekime Analytics'']] at the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* March 7, 8, and 21, 2022: Ivo Dinov is organizing a [[SOCR_News_ISI_DSPA_Training_2022 | 2022 ISI Short Course on Data Science and Predictive Analytics (DSPA)]], distance (virtual) event.&lt;br /&gt;
* March 13, 2022: Ivo Dinov is organizing a [[SOCR_News_APS_GDS_ShortCourse_March_2022| day-long short course ''Longitudinal Data Tensor-Linear Modeling and Space-kime Analytics'']] for the [https://www.aps.org/meetings/ March 2022 APS Meeting].&lt;br /&gt;
* January 21, 2022: [[SOCR_News_SOCR_Winter2022_retreat | SOCR Winter 2022 Retreat]] (hybrid; in-person + ZM), 5:00-6:30PM US ET (GMT-5).&lt;br /&gt;
* January 6, 2022: [https://mdp.engin.umich.edu/mdp_events/all/ 2022: SOCR-MDP 2022 JumpStart Event], Time: 6:00 -8:30 pm, Box-Dinner pick up: 6:00 pm – 6:30 pm Duderstadt Atrium, SOCR Group meets at [https://maps.studentlife.umich.edu/building/chrysler-center-continuing-engineering-education Chrysler Building CHY #165], 6:15-7:30 pm.&lt;br /&gt;
&lt;br /&gt;
==2021==&lt;br /&gt;
* November 12-13, 2021: [https://www.socr.umich.edu/people/SOCR_Personnel_Yueyang.html Yueyang Shen] presented a [https://meetings.aps.org/Meeting/EGLSF21/Session/J02.7 talk] entitled [https://wiki.socr.umich.edu/images/1/1e/YueyangShen_APS_WDE_Spacekime_Fall_2021_Talk.pdf Wheeler Dewitt equation in spacekime] at the [https://meetings.aps.org/Meeting/EGLSF21/APS_epitome Fall 2021 APS/EGLS meeting].&lt;br /&gt;
* November 6-10, 2021: Ivo Dinov was inducted as an [https://www.sigmanursing.org/connect-engage/news-detail/2021/08/11/sigma-announces-2021-international-award-recipients-and-honorary-members ''Sigma Honorary Member''] at the [https://www.sigmanursing.org/connect-engage/meetings-events/convention 46th Biennial Convention], Indianapolis, Indiana, USA.&lt;br /&gt;
* October 27-28, 2021: Ivo Dinov presented [https://wiki.socr.umich.edu/images/0/03/Dinov_UMich_BigHealthDataPillar_MBDH_Conf_2021.pdf ''MBDH Big Health Data Pillar: Open Biomedical Team Science''] at the [https://midwestbigdatahub.org/2021-rcm-agenda/ 2021 Midwest Big Data Hub Symposium].&lt;br /&gt;
* August 30 – September 4, 2021: Jared Tianyi Chai, Mark Bobrovnikov, and Ivo Dinov are presenting [[SOCR_News_IASE_Distributome_2021 | Probability Distributome – Computing, Visualization, and Instruction]] at the 2021 IASE Conference on Statistics Education in the Era of Data Science. &lt;br /&gt;
* August 27, 2021: [[SOCR_News_SOCR_Fall2021_retreat | SOCR Fall 2021 Retreat]], 8:00-9:30 AM US ET (GMT-4).&lt;br /&gt;
* August 25, 2021: Ivo Dinov presented [[SOCR_News_Bioinfo504_RCR_2021| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the 2021 Rigor and Transparency to Enhance Reproducibility Workshop, University of Michigan.&lt;br /&gt;
* July 26-30, 2021: Kayvan Najarian, Nambi Nallasamy, Ivo Dinov, Michael Mathis, Ryan Stidham, Jonathan Gryak, Michael Sjoding are organizing the [[SOCR_News_MIDAS_Biomedical_Bootcamp_2021| 2021 MIDAS Data Science for Biomedical Scientists Bootcamp]], distance (virtual) event.&lt;br /&gt;
* July 2021 - June 2026: SOCR investigators led the establishment of an innovative T32 NIH-NIGMS Doctoral Training Program - [https://bids-tp.umich.edu/ the University of Michigan Biomedical Informatics and Data Science Training Program (BIDS-TP)].&lt;br /&gt;
* June 21, 2021 (8 AM US ET, GMT-4): Ivo Dinov presented [[SOCR_News_SOCR_D43_2021_SummerSchool | Data Science, Analytics, and AI in Health]], at the D43 Summer School.&lt;br /&gt;
* June 16-17, 2021: Ivo Dinov organized a [[SOCR_News_ISI_WSC_DSPA_Training_2021| 2021 ISI/WSC Training and Education Bootcamp on ''Data Science and Predictive Analytics (DSPA)'']], distance (virtual) event.&lt;br /&gt;
* April 10, 2021: Ivo Dinov is presenting [[SOCR_News_APS_Ohio_Spacekime_2021| ''Data Science, Time Complexity, and Spacekime Analytics'']], at the 2021 Meeting of the American Physical Society (APS) Ohio-Region Section, remote distance event.&lt;br /&gt;
* March 30, 2021: Ivo Dinov was interviewed by [https://www.jerseyindie.com/philip-perry Phil Perry] for a [https://bigthink.com/hard-science/ BigThink Hard Science] article [https://bigthink.com/hard-science/spacekime-theory/ 'Spacekime theory' could speed up research and heal the rift in physics].&lt;br /&gt;
* March 19, 2021 (12PM ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_OpenScience_DataSharing_UVA_Mar2021.pdf ''Big Neuroscience, Data Sharing &amp;amp; Predictive Health Analytics''] at the [http://innovation.lab.virginia.edu UVA Biomedical Data Science Innovation Seminar Series], [https://www.youtube.com/watch?v=qukLnbIa1qk video stream/archive].&lt;br /&gt;
* March 5, 2021 (9 ET, GMT-5): Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/Dinov_UM_Psych_Neuroimaging_2021.pdf Open Computational Neuroscience: Research, Development &amp;amp; Training] at the University of Michigan [https://www.umich.edu/~nii/ NII-Methods seminar series], [https://umich-health.zoom.us/j/99279105041?pwd=SFIrTFFtTlpodHZzQjhWNWhSZnpnQT09 Zoom link].&lt;br /&gt;
* February 23, 2021: Ivo Dinov presented [https://socr.umich.edu/docs/uploads/2021/DigitalTransformation_LHS_SoD_UMich_2021_Dinov.pdf ''Data De‐Identification &amp;amp; Clinical Decision Support''] at the [https://medicine.umich.edu/dept/lhs/service-outreach/learning-health-system-collaboratory 2021 Data Standards and Learning Health Systems–Challenges and Opportunities Symposium], Ann Arbor, MI. [https://umich.zoom.us/webinar/register/WN_0XSNkkotQTKQfTAAv-XJvQ Registration] and [https://umich.zoom.us/j/99190944947 1:15 PM ET special session 1; ZOOM access], [https://youtu.be/U3CLm25HKdE?t=1420 recorded video].&lt;br /&gt;
* February 04, 2021: Ivo Dinov presented [[SOCR_News_MICDE_Seminar_2021| ''Data Science, Time Complexity, and Spacekime Analytics'', MICDE Seminar Series]], Ann Arbor, MI.&lt;br /&gt;
* January 19, 2021: [[SOCR_News_SOCR_Winter2021_retreat | Virtual SOCR Winter 2021 Retreat]], 5:00-6:30 PM US ET (GMT-4).&lt;br /&gt;
* January 6-9, 2021: Ivo Dinov organized a [[SOCR_News_JMM_DC_Session_2021| 2021 JMM Special Session on ''Foundations of Data Science: Mathematical Representation, Computational Modeling, and Statistical Inference'']], Washington, DC.&lt;br /&gt;
* January-April 2021: [https://umma.umich.edu/curriculum-collection/health-sciences-650 University of Michigan Museum of Art (UMMA) Curriculum Collection: Blending, Arts, Sciences, and Health Analytics (HS 650; Data Science and Predictive Analytics )].&lt;br /&gt;
&lt;br /&gt;
==2020==&lt;br /&gt;
* November 2020: [https://umich.instructure.com/courses/387294/pages/amia-video-data-tools-services-nov-2020 Data Resources, Analytical Tools &amp;amp; Cloud Services for Clinical Decision Support (video)], [https://socr.umich.edu/docs/uploads/2020/AMIA_2020_Dinov.pdf Slidedeck] (AMIA 2020).&lt;br /&gt;
* November 10-11, 2020: Ivo Dinov is presenting [[SOCR_News_MIDAS_Symposium_2020| ''Computational Neuroscience, Time Complexity, and Spacekime Analytics'' at the 2020 Data Science Annual Symposium]], Ann Arbor, MI.&lt;br /&gt;
* October 8, 2020: Ivo Dinov is presenting [[SOCR_News_DCMB_ToolsTechSeminar_2020| ''Spacekime Analytics'' at the DCMB Tools and Technology Seminar]], Ann Arbor, MI.&lt;br /&gt;
* September 13, 2020: Ivo Dinov is presenting [[SOCR_News_MDS_2020_BigData | ''What is Big Neuro Data? Where is it? How to Use it? Why is it Important?'' at the 2020 Movement Disorders Society Symposium's Special Topics Session (601: Big Data Analytics in Clinical Research for Movement Disorders)]].&lt;br /&gt;
* September 11, 2020: Ivo Dinov is presenting [[SOCR_News_Biophysics_2020_Spacekime | Data Science, Time Complexity, and Spacekime Analytics]] at University of Michigan Biophysics.&lt;br /&gt;
* September 4, 2020: Ivo Dinov is presenting the [http://neurosciencenetwork.org/ACNN_Workshop_2020.html SOCR and DSPA resources at the 2020 ACNN Symposium on Big Neuroscience Data].&lt;br /&gt;
* August 28, 2020: [[SOCR_News_SOCR_Fall2020_retreat | Virtual SOCR Fall 2020 Retreat]], 8 AM US ET (GMT-4).&lt;br /&gt;
* August 26, 2020: Ivo Dinov is presenting [[SOCR_News_Bioinfo504_RCR_2020| Data Sharing &amp;amp; Open, Rigorous, Reproducible Science]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 2020 Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* June 23-25, 2020: Ivo Dinov is presenting &amp;quot;''The Interface between Data Science and Health Analytics''&amp;quot; at the [[SOCR_News_MIWI_SummerSchool_2020 |MIWI (Michigan Integrative Well-Being and Inequality) Summer School]], Ann Arbor, MI. &lt;br /&gt;
* June 1-2, 2020: Ivo Dinov is presenting [[SOCR_News_NIH_BRAIN_2020| &amp;quot;Data Science, Time Complexity, and Spacekime Analytics&amp;quot; at the 2020 6th annual BRAIN Initiative Investigators Virtual Meeting]].&lt;br /&gt;
* May 27, 2020: Ivo Dinov is presenting [[SOCR_News_Neuromatch2.0_2020| &amp;quot;''Computational Neuroscience, Time Complexity, and Spacekime Analytics''&amp;quot; at Neuromatch 2.0 Unconference]].&lt;br /&gt;
* February 14, 2020: Ivo Dinov is presenting [[SOCR_News_2020_UM_AIM_Seminar_Spacekime_Dinov | Data Science, Time Complexity and Spacekime Analytics]] at the University of Michigan Applied Interdisciplinary Mathematics (AIM) Seminar Series.&lt;br /&gt;
* January 9, 2020: [[SOCR-MDP_2020_AllHandsJumpStartMeeting | 2020 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2019==&lt;br /&gt;
* December 3, 2019: Ivo Dinov is presenting [[SOCR_News_UMich_SPH_MLEED_2019 | SOCR DataSifter: A Statistical Obfuscation Technique enabling Effective Data Sharing, at the University of Michigan SPH Environmental Epidemiology Seminar Series]]. &lt;br /&gt;
* October 16-17, 2019: Ivo Dinov is presenting at the [[SOCR_News_BrockU_Symposium_2019 |BIG DATA Analytics in Health Care Session of Re-Imaging Health Symposium, Brock University, St. Catharines, ON L2S 3A1, Canada]]. &lt;br /&gt;
* October 11-12, 2012: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/f/f0/NGP_2019_Dinov_DataBlitz_DataNeuroscience_2019.pdf  Big Data Neuroscience, Computational Methods, &amp;amp; Predictive Analytics] at the 2019 Neuroscience Graduate Program Retreat (Faculty Data Blitz), Roscommon, Michigan.&lt;br /&gt;
* Sept 26, 2019: Ivo Dinov is presenting a talk [[SOCR_News_SpacekimeAnalytics_Fall2019 | Longitudinal Spacekime Analytics: Time Complexity &amp;amp; Inferential Uncertainty]] ([https://sph.umich.edu/biostat/events.php University of Michigan, SPH/Biostatistics], 3:30-4:30 PM, 3755 SPH I).&lt;br /&gt;
* Sept 19-20, 2019: Ivo Dinov and the Advanced Computational Neuroscience Network (ACNN) are organizing the [http://www.neurosciencenetwork.org/ACNN_Workshop_2019.html 4th annual (2019) Big Data Neuroscience Workshop, University of Michigan, Ann Arbor, MI]. &lt;br /&gt;
* Sept 3, 2019: [[SOCR_News_SOCR_Fall2019_retreat | SOCR Fall Retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* August 29, 2019 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2019 | Introduction to UM Health Analytics, SOCR, MIDAS, NGP]].&lt;br /&gt;
* August 18–23, 2019: Ivo Dinov is organizing an [[SOCR_News_ISI_WSC_IPS35_2019| Invited Presenter Session; IPS35: Imaging Statistics and Predictive Data Analytics]] at the [https://isi-web.org International Statistical Institute]’s [https://isi-web.org/index.php/activities/world-statistics-congresses 2019 World Stats Congress].&lt;br /&gt;
* July 23, 2019: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2019/Dinov_ML_BreastCancer_2019.pdf Breast Cancer Risk Prediction using Model-based and Model-Free Techniques] at the [https://www.rogelcancercenter.org University of Michigan Rogel Cancer Center], Cancer Center #6317.&lt;br /&gt;
* May 9, 2019: Ivo Dinov is presenting [[SOCR_News_ICPSR_PIBS_2019| Open Science and Data Sharing]] at the [https://www.icpsr.umich.edu/icpsrweb/sumprog/courses/0279 ICPSR Rigor and Transparency to Enhance Reproducibility Workshop], University of Michigan.&lt;br /&gt;
* April 12-13, 2019: Ivo Dinov is presenting [[SOCR_News_FSU_DataImpact_Symposium_2019 | DataSifter: Sharing of Sensitive Data via Statistical Obfuscation]] at the 2019 [https://ani.stat.fsu.edu/60th/ Statistics, the Impact of Big Data Conference, FSU, Tallahassee, FL].&lt;br /&gt;
* March 26, 2019: Ivo Dinov is presenting [[SOCR_News_AA_ASA_March_2019| Challenges and Opportunities in Predictive Big Data Analytics]] at the Ann Arbor Chapter of ASA.&lt;br /&gt;
* February 23, 2019 (9:30AM - 5 PM): Alex Kalinin is organizing the 4-th annual [https://alxndrkalinin.github.io/a2-dlearn-4/ 2019 Ann Arbor Deep Learning Event (a2-dlearn4)], UM College of Engineering Francois-Xavier Bagnoud Building (1109 1320 Beal Ave., Ann Arbor, MI)&lt;br /&gt;
* January 10 and January 15, 2019: [[SOCR-MDP_2019_AllHandsJumpStartMeeting | 2019 SOCR-MDP All-Hands JumpStart Meeting]].&lt;br /&gt;
&lt;br /&gt;
==2018==&lt;br /&gt;
* November 2, 2018: Ivo Dinov is presenting on [[SOCR_News_2018_NIAAA_SfN_SD | Open Data Science and Predictive Health Analytics]] at the SfN 2018.&lt;br /&gt;
* Oct 25, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_DataSifter_UMich_SUMIT_2018_short.pdf DataSifter: Sharing of sensitive information via statistical obfuscation]'' ([https://www.youtube.com/watch?v=SIp81fe8IWQ Video]) at the [https://www.safecomputing.umich.edu/events/sumit/2018 14&amp;lt;sup&amp;gt;th&amp;lt;/sup&amp;gt; annual cyber security conference on Security at University of Michigan IT (SUMIT)], an event for National Cybersecurity Awareness Month on the latest technical, legal, policy, and operational trends, threats, and tools in cybersecurity and privacy.&lt;br /&gt;
* Oct 15-16, 2018: Ivo Dinov is presenting the [http://socr.umich.edu/docs/uploads/2018/CNSECCS_MIDAS_Dinov_2018.pdf Michigan Institute of Data Science: Computational Challenges and Research Opportunities] at the [https://micde.umich.edu/centers/cnseccs/symposia/ 2018 Center for Network and Storage-Enabled Collaborative Computational Science (CNSECCS) Symposium], Ann Arbor, Michigan.&lt;br /&gt;
* Oct 12, 2018: [[SOCR_News_2018_MNORC_SOCR_HAC_Workshop|MNORC-IBIC/SOCR/HAC Health Data Analytics Workshop]].&lt;br /&gt;
* Sept 21, 2018: Ivo Dinov is presenting [https://midas.umich.edu/event/midas-seminar-series-presents-ivo-d-dinov-phd-university-of-michigan/ ''The Enigmatic Kime: Time Complexity in Data Science''] at the [https://midas.umich.edu/seminar-series/ University of Michigan Institute for Data Science (MIDAS) Seminar Series], [http://socr.umich.edu/docs/uploads/2018/Dinov_TCIU_Kime_MIDAS_2018.pdf Slidedeck], [https://www.youtube.com/watch?v=yivkoU51MPM YouTube video of this seminar].&lt;br /&gt;
* Sept 18, 2018 : [[SOCR_News_SOCR_Fall2018_retreat | SOCR semi-annual retreat]] (12-2 PM, SNB 1250).&lt;br /&gt;
* Sept 15, 2018: Ivo Dinov is presenting ''[http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveHealthAnalytics_2018_DCMB.pdf High-Dimensional Biomedical Data &amp;amp; Predictive Health Analytics]'' at the Fall 2018 [https://medicine.umich.edu/dept/computational-medicine-bioinformatics DCMB/UMich] Retreat, Frankenmuth, MI.&lt;br /&gt;
* Sept 6, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/2018/Dinov_PredictiveNeuroAnalytics_2018_ACNN.pdf ''Data Science &amp;amp; Predictive Neuro-Analytics''] at the [http://www.neurosciencenetwork.org/ACNN_Workshop_2018.html 2018 Advanced Computational Neuroscience Network (ACNN) Workshop], Case Western Reserve University, Ohio.&lt;br /&gt;
* August 30, 2018 (11:20 AM): [https://nursing.umich.edu/about/events-calendar New School of Nursing Faculty Orientation]: Ivo Dinov will present an [[SOCR_Intro_UMich_Fall_2018 | Introduction to SOCR]].&lt;br /&gt;
* July 26, 2018: Ivo Dinov is presenting [http://wiki.socr.umich.edu/images/0/09/Dinov_PredictiveDataAnalytics_2018_UMich_Biostats.pdf Data Science &amp;amp; Predictive Health Analytics] at the [https://sph.umich.edu/bdsi/about/symposium.html 2018 Symposium on Big Data, Human Health and Statistics, University of Michigan].&lt;br /&gt;
* June 22, 218: Ivo Dinov presented the [http://neurosciencenetwork.org Advanced Computational Neuroscience Network (ACNN)] at the [https://www.bi.vt.edu/nsf-big-data-2018/schedule-joint-pi-meeting-2018 NSF BIGDATA Conference in Washington DC].&lt;br /&gt;
* May 17, 2018, Ivo Dinov is presenting a day-long series on ''Big Data &amp;amp; Health Analytics'' at the [http://nursing.umich.edu/academics/global-sexual-health-summer-institute Global Sexual Health Summer Institute]. Specifically, Dr. Dinov is presenting an interactive session on [[SOCR_News_2018_UMSN_SummerInstitute| Big Healthcare Data: Research Challenges, AI Capabilities, and Educational Opportunities]].&lt;br /&gt;
* May 11, 2018, Ivo Dinov is presenting [[SOCR_News_UMich_CCC_HnN_Retreat_2018| Exploratory, Confirmatory, &amp;amp; Predictive Big Cancer Data Analytics]] at the [https://www.mcancer.org/head-and-neck-cancer/research University of Michigan Rogel Cancer Center Head and Neck Cancer Retreat].&lt;br /&gt;
* May 04, 2018: Ivo Dinov is presenting [http://socr.umich.edu/docs/uploads/Dinov_HealthAnalytics_2018_InnovationBoard.pdf Predictive Health Analytics] at the University of Michigan Nursing Board of Science Innovation Forum.&lt;br /&gt;
* April 25, 2018: Ivo Dinov is presenting [[SOCR_Intro_UMich_2018 | Introduction to SOCR]] at the [http://nursing.umich.edu/about/departments Health Behavior and Biological Sciences Department]&lt;br /&gt;
* April 24, 2018: Ivo Dinov is presenting two talks at the [https://www.binghamton.edu/transdisciplinary-areas-of-excellence/data-science/ Data Science Initiative, SUNY Binghamton]: [http://socr.umich.edu/docs/uploads/MIDAS_Dinov_2018.pdf Michigan Institute of Data Science – Organization, Education Challenges and Research Opportunities] and [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2018_SUNY_B.pdf Compressive Big Data Analytics (CBDA)].&lt;br /&gt;
* April 18, 2018: SOCR semi-annual retreat (12-2 PM, SNB 1250).&lt;br /&gt;
&lt;br /&gt;
==2017==&lt;br /&gt;
* Dec 08, 2017, (12 Noon ET, GMT-5) Ivo Dinov is presenting &amp;quot;Big Brain Data Science &amp;amp; Predictive Health Analytics&amp;quot; webinar at the [https://bigdatau.ini.usc.edu/data-science-seminars BD2K Guide to the Fundamentals of Data Science Series], [https://attendee.gotowebinar.com/register/7111730654574370307 Webinar ID: 932-094-291], [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_BD2K_TCC.pdf Slides].&lt;br /&gt;
* Nov 05, 2017, A. Sharma (Emory University), W. Hsu (UCLA), E. Siegel (University of Maryland), K. Cheng (Penn State University) and I. Dinov (University of Michigan) are organizing a half-day workshop at [https://www.amia.org/amia2017 AMIA 2017]. The title of the workshop is [https://www.amia.org/amia2017/workshops (W15) Does Integrative Data Analytics on Biomedical Imaging Bring Us Closer to Precision Medicine?], which is sponsored by the [https://www.amia.org/programs/working-groups/biomedical-imaging-informatics AMIA Biomedical Imaging Working Group], [http://www3.hilton.com/en/hotels/district-of-columbia/washington-hilton-DCAWHHH/index.html Washington Hilton, 1919 Connecticut Ave., NW, Washington, DC 20009]. Dr. Dinov's keynote lecture is on [http://socr.umich.edu/docs/uploads/Dinov_PredictiveDataAnalytics_2017_AMIA_BIWG.pdf Big Brain Data &amp;amp; Predictive Analytics].&lt;br /&gt;
* Sept 8-9, 2017, Ivo Dinov presents ''Predictive Big Brain Data Analytics'' at the [http://neurosciencenetwork.org/ACNN_Workshop_2017.html 2017 Advanced Computational Neuroscience Network (ACNN) Big Data Workshop], Indiana University, Bloomington, IN. &lt;br /&gt;
* Aug 15, 2017, Ivo Dinov is presenting a keynote lecture on [[SOCR_News_IBRO_INSF_Malaysia_2017#Aug_15-16.2C_2017_3rd_Malaysia_Telemedicine_Conference_.282017.29 | Predictive Data Analytics]] at the [http://med.monash.edu.my/campaign/telemed/index.html Malaysia Telemedicine Conference 2017] (August 15-17, 2017, Kuala Lumpur, Malaysia).&lt;br /&gt;
* Aug 10-18, 2017: [http://www.umich.edu/~dinov/ Ivo Dinov], [http://www.med.monash.edu.au/psych/bmh/people/alex.html Alex Fornito], [https://findanexpert.unimelb.edu.au/display/person24599 Andrew Zalesky], [http://gablab.mit.edu/index.php/14-sample-data-articles/157 Satrajit Ghosh], and [https://www.researchgate.net/profile/Eric_Tatt_Wei_Ho Eric Tatt Wei Ho] are organizing a INCF/IBRO Neuroscience Summer School for graduate students and postdoctoral fellows. [[SOCR_News_IBRO_INSF_Malaysia_2017 | Dr. Dinov will be lecturing on (1) Statistical Computing, (2) High-Throughput Processing of Big Neuroscience Data, and (3) Neuroimaging-genetics]]. The Summer Neuroscience School is part of the [https://www.incf.org International Neuroinformatics Coordinating Facility (INCF)]/[http://ibro.info/ International Brain Research Organization] [http://ibro.info/events/applications-open-for-ibro-aprc-school-on-neuroinformatics-and-brain-network-analysis/ IBRO-APRC School on Neuroinformatics and Brain Network Analysis], Kuala Lumpur, Malaysia. &lt;br /&gt;
* July-September 2017, [http://dspa.predictive.space Summer 2017: Data Science and Predictive Analytics (UMich HS650)], a Massive Open Online Course (MOOC).&lt;br /&gt;
* May 8-12, 2017: Ivo Dinov is offering a [http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ week-long Bootcamp: Predictive Big Data Analytics using R]. This deep dive into modern Big Data Analytics will take place Monday (May 8, 2017) through Friday (May 12, 2017), 8:00AM to 4:00PM, at SNB 1250, University of Michigan, 426 North Ingalls, Ann Arbor, MI. The event will also be ''streamed live'' ([http://www.socr.umich.edu/people/dinov/2017/Spring/PBDA_R_Bootcamp/ see details on the bootcamp site]).  &lt;br /&gt;
* March 23, 2017: FireFox 52+ will not support Java plug ins. SOCR users that are interested in continuing to access SOCR Java Applets via FireFox need to install the 32-bit version of [https://support.mozilla.org/t5/Problems-with-add-ons-plugins-or/Why-do-Java-Silverlight-Adobe-Acrobat-and-other-plugins-no/ta-p/31069 FF Extended Support Release (52 ESR)].&lt;br /&gt;
* Mar 16-17, 2017: The principals (Rich Gonzales, Ivo Dinov, John Marcotte, Franco Pestilli, Olaf Sporns, Andrew Saykin, Dhabaleswar Panda, Xiaoyi Lu, Hari Subramoni, Satya Sahoo, Daniel Marcus, and Lei Wang) of the [http://neurosciencenetwork.org/ Advance Computational Neuroscience Network (ACNN)] will give talks and present posters at the [https://www.bi.vt.edu/nsf-big-data 2017 NSF BigData Meeting] at the Omni Shoreham Hotel, Washington DC.&lt;br /&gt;
* Mar 3, 2017: Ivo Dinov is presenting [http://socr.umich.edu/HTML5/BrainViewer Brain Visualization], [http://pipeline.loni.usc.edu High-throughput computational processing], and [http://socr.umich.edu/HTML5/SOCRAT the SOCRAT Statistical Computing Framework] at the [https://sites.google.com/a/umich.edu/brainhack-global-a2-2017/ 2017 Brainhack-Global Meet in Ann Arbor].&lt;br /&gt;
&lt;br /&gt;
==2016==&lt;br /&gt;
* Nov 12, 2016: Alex Kalinin is co-organizing [http://bit.ly/a2-dlearn-2016 a2-dlearn 2016 – the 2nd annual Ann Arbor Deep Learning event] and will present ''Deep Learning Year in Review 2016: Computer Vision Perspective''. The goal of '''a2-dlearn 2016''' is to bring together deep learning enthusiasts, researchers and practitioners from a variety of backgrounds. Venue: 1670 Bob and Betty Beyster Building, University of Michigan, 2260 Hayward St, Ann Arbor, MI.&lt;br /&gt;
* Nov 01, 2016: Ivo Dinov is presenting Predictive Big Data Analytics Workshop, School of Nursing University of Michigan Analytics Seminar Series, 1:00PM, 1240 SNB. This presentation will focus on Predictive Big Data Analytics. We'll go over characteristics and fundamentals of Big Data, methodological and computational challenges, health science research, and opportunities. Applications to biosocial (Medicare/Economics) and neurodegenerative disorders (Parkinson's Disease) will be presented. The foundations of compressive Big Data Analytics (CBDA) and several demos will be shown.&lt;br /&gt;
* In October 2016, Ivo Dinov participated in an [http://zika.smartercrowdsourcing.org/anlisis-predictivo-conference.html expert International panel including government officials of a dozen countries affected by the Zika virus]. The panel reviewed evidence and made recommendations to public officials on strategies to combat the spread of the Zika virus and the associated microcephaly in newborns. &lt;br /&gt;
* Sept 20-21, 2016:  Ivo Dinov, Rich Gonzales, George Alter (University of Michigan), Franco Pestilli, Olaf Sporns, Andrew Saykin (Indiana University), Dhabaleswar Panda, Khaled Hamidouche, Xiaoyi Lu, Hari Subramoni (OSU), Satya Sahoo (CWRU), Daniel Marcus (Washington University), and Lei Wang (Northwestern University) are organizing a 2-day [http://www.neurosciencenetwork.org/ACNN_Workshop_2016.html Midwest Workshop on Big Neuroscience Data, Tools, Protocols &amp;amp; Services], Ann Arbor, MI.&lt;br /&gt;
* Aug 04, 2016 (8:30-10:30 AM): Ivo Dinov is presenting [http://www.amstat.org/meetings/jsm/2016/onlineprogram/AbstractDetails.cfm?abstractid=319094 Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities], at the [http://www.amstat.org/meetings/jsm/2016/onlineprogram/ActivityDetails.cfm?SessionID=212981 Recent advances in massive imaging data analysis, Section on Statistics in Imaging], at the [http://www.amstat.org/meetings/jsm/2016/ Joint Statistics Meeting (JSM)], Chicago, IL, (CC-W McCormick Place Convention Center, West Building, Session: 659, Thu, 8/4/2016, 8:30 AM - 10:20 AM, '''CC-W187b''').&lt;br /&gt;
* July 08, 2016: Ivo Dinov was [http://www.pbs.org/wgbh/nova/next/body/theres-hope-for-fmri-despite-major-software-flaws/ interviewed by NOVA (WGBH/PBS)] on a [http://www.pnas.org/content/early/2016/06/27/1602413113 recent PNAS Report identifying significant potential shortfalls of Big Data functional magnetic resonance imaging (fMRI) studies], of which there may be over 35,000 in the past quarter century. The article used 500 normal control subjects (null data) to generate 3 million simulation studies where every experiment included randomly chosen subjects, either resting state or task activation fMRI, and found false-positive discoveries (significant grouping effects where there were none) up to 70% of the simulations. Although this does not discredit any specific previously published fMRI findings, the investigations suggests the need for novel Big Data analytics methods, and scalable software tools, that can reduce the false-positive rate.&lt;br /&gt;
* May 19, 2016 (12:00 PM - 1:00 PM EDT): Ivo Dinov is presenting an [https://www.amia.org/education/webinars AMIA Education &amp;amp; Training Webinar] entitled [http://www.amia.org/education/webinars/predictive-big-data-analytics-imaging-genetics-fundamentals-research-challenges Predictive Big Data Analytics: Imaging-Genetics Fundamentals, Research Challenges, and Opportunities], [http://wiki.socr.umich.edu/images/e/ef/Dinov_PredictiveBigDataAnalytics_2016_BMII_WG_2016.pdf PDF Slides] and [https://knowledge.amia.org/webinars/webinars-working-group-1.1423688/2016-webinars-1.2888451/2016-webinars-1.2888452/bmii-wg-predictive-big-data-analytics-1.3109533/bmii-wg-predictive-big-data-analytics-1.3109534 video/webcast].&lt;br /&gt;
* Mar 07, 2016, Ivo Dinov was [https://www.isi-web.org/images/news/ISI_Memb%20Elect%202015-2016_News%20Item.pdf Elected as a Member] of the [http://isi-web.org International Statistical Institute (ISI)]. ISI is the world’s most prestigious Statistical Organization Established in 1885 as a non-profit, non-government, organization with a consultative status within the Economic and Social Council of the United Nations. The ISI organizes the [http://www.isi2017.org World Statistics Congresses] every two years across the Globe.&lt;br /&gt;
* Jan 19, 2016, Ivo Dinov gave a lecture on ''Big, Deep, and Dark Data: Fundamentals, Research Challenges, and Opportunities'' at the [http://www.src.isr.umich.edu Survey Research Center (SRC)] seminar series, Institute for Social Research, University of Michigan.&lt;br /&gt;
&lt;br /&gt;
==2015==&lt;br /&gt;
* Nov 23-24, 2015, Ivo Dinov is giving a keynote address ([http://socr.umich.edu/docs/uploads/Dinov_BigDeepDarkDataAnalytics_2015_MonterreyTech.pdf Predictive Big Data Analytics: Using Large, Complex, Heterogeneous, Incongruent, Multi-source and Incomplete Observations to Study Neurodegenerative Disorders]) at the [http://big-data.csf.itesm.mx/~BigData/pass/Evento/en 2015 Big Data Analytics Experience Conference], Tecnológico de Monterrey (Monterrey Tech), Santa Fe, 01389 Ciudad de México, D.F., Mexico.&lt;br /&gt;
* Oct 14, 2015, Ivo Dinov interviewed by ''Inside Higher Ed'' on [http://www.insidehighered.com/news/2015/10/13/colleges-explain-why-they-double-dipped-moocs 'Double-Dipping' With MOOCs].&lt;br /&gt;
* Oct 9-10, 2015, Carl Kesselman and Ivo Dinov organized a [http://bd2k.ini.usc.edu/events/ working group on Scientific Tools and Workflows] for Big Data Discovery Science, Palm Springs, CA.&lt;br /&gt;
* Oct 06, 2015, Ivo Dinov presented the [http://midas.umich.edu/symposium/ Michigan Institute for Data Science (MIDAS) Education and Training Program].&lt;br /&gt;
* Oct 02, 2015, Petros Petrosyan, Sam Hobel, and Ivo Dinov present the [http://bd2k.ini.usc.edu/resources/documents/Pipeline_Demo_Day_Booklet.pdf Hippocampal Meta-Analysis Workflow] at USC Pipeline Demo Day.&lt;br /&gt;
* Sept 23, 2015, Ivo Dinov presented [http://events.umich.edu/event/24584 Exploratory Big Data Analytics] at the University of Michigan ([http://wiki.socr.umich.edu/images/a/ab/Dinov_SciViz_CSCD_Seminar_2015.pdf PDF]).&lt;br /&gt;
* Sept 10 and Sept 17, 2015, Ken Powell, Eric Michielssen, and Ivo Dinov presented opportunities for [http://arc.umich.edu/2015/08/27/info-sessions-graduate-studies-in-computational-and-data-sciences-at-u-m-sept-10-17 graduate studies in computational and data sciences at the University of Michigan].&lt;br /&gt;
* August 8–13, 2015, Ivo Dinov organized a special session on [[SOCR_Events_JSS_2015|Big Data: Modeling, Tools, Analytics, and Training]] at the [http://www.amstat.org/meetings/jsm/2015 2015 Joint Statistical Meeting], Washington State Convention Center, 800 Convention Place, Seattle, WA 98101. &lt;br /&gt;
* June 14-18, 2015, Ivo Dinov and his colleagues presented &amp;quot;The Pipeline Environment: A Scalable, Distributed and Service-Oriented Neuroimaging and Genetics&amp;quot; at the [http://ohbm.loni.usc.edu 21st Annual Meeting of the Organization for Human Brain Mapping], Honolulu, Hawaii.&lt;br /&gt;
* June 14-17, 2015, Ivo Dinov presented the [[SOCR_Events_SSC_2015|SOCR Resources]] at the [http://www.ssc.ca/en/meetings/2015 43rd Annual Meeting of the Statistical Society of Canada (SSC)], [http://www.dal.ca/ Dalhousie University], Halifax, NS, Canada.&lt;br /&gt;
* May 5, 2015, Deb Barton and Ivo Dinov presented the [http://www.socr.umich.edu/CSCD/ Center for Complexity and Self-management of Chronic Disease (CSCD)] at the [http://www.ninr.nih.gov/newsandinformation/newsandnotes 2015 National Institute of Nursing Research Center Directors Meeting], National Institutes of Health, Bethesda Maryland, Natcher Conference Center: Room E1/E2.&lt;br /&gt;
* April 21-24, 2015, Ivo Dinov organized a [http://midas.umich.edu/event/micro-big-data-analytics-workshop-2015/ micro Big Data Analytics workshop at the University of Michigan Institute for Data Science (MIDAS)]. The focus of this micro-workshop is to lay down the foundation for developing a new Compressive Big Data Analytics (CBDA) foundation enabling representation, modeling, analysis and interrogation of large, incongruent multi-source, incomplete and messy data. The highlights of the workshop include talks from [http://www.socr.umich.edu/CSCD/html/events/Amiri_CSCD_Presents_2015.html Dr. Saeid Amiri] and [http://www.socr.umich.edu/CSCD/html/events/Ahmed_MIDAS_Presents_2015.html Dr. S. Ejaz Ahmed].  &lt;br /&gt;
* February 18, 2015, Ivo Dinov gave a lecture on &amp;quot;Management, Modeling &amp;amp; Analytic Challenges of Big Biomedical Data&amp;quot; at the [http://www.psych.med.umich.edu/events/579 University of Michigan Psychiatry Grand Rounds].&lt;br /&gt;
* January 13-14, 2015, Ivo Dinov presented the &amp;quot;Pipeline graphical workflow environment for computational genomics, proteomics, image and shape analysis&amp;quot; at the [http://bd2k.ini.usc.edu Big Data Discovery Science (BDDS)] Meeting in Seattle, WA.&lt;br /&gt;
&lt;br /&gt;
==2014==&lt;br /&gt;
* Aug 05, 2014, [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting the new [[SOCR_Events_SMHS_2014|UMSN Analytics Curriculum (Scientific Methods for Health Sciences, SMHS)]].&lt;br /&gt;
* July 8-9, 2014: [http://www.umich.edu/~dinov/ Ivo Dinov] is presenting &amp;quot;Portable Pipeline Workflows for Nuclear and Chromosomal Shape Morphometry Analysis&amp;quot; at the [http://atheylab.ccmb.med.umich.edu/4d-nucleome-workshop 4D Nucleome Workshop, University of Michigan].&lt;br /&gt;
* May 22, 2014, Ivo Dinov (SOCR/Michigan), Dennis Pearl (MBI/OSU), and Kyle Siegrist (VLPS/UAH) are organizing a Workshop entitled [http://www.distributome.org/meetings/eCOTS_2014/ Web Resources for Interactive Probability Instruction] at the [http://www.causeweb.org/ecots/ 2014 eCOTS (Electronic Conference On Teaching Statistics)] conference.&lt;br /&gt;
* April 28, 2014: Ivo Dinov is interviewed on &amp;quot;[http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov The Big Data]&amp;quot; by Center for Science, Technology, and Security Policy (CSTSP), at the American Association for the Advancement of Science (AAAS).&lt;br /&gt;
* April 1-2, 2014: Ivo Dinov presented a talk entitled &amp;quot;Big Data Challenges: Data Management, Analytics &amp;amp; Security&amp;quot; at the AAAS/FBI [http://www.aaas.org/event/big-data-life-sciences-and-national-security Summit on Big Data, Life Sciences, and National Security] in Washington DC. See [http://www.aaas.org/news/big-data-blog-part-v-interview-dr-ivo-dinov blog post].&lt;br /&gt;
* Jan 30, 2014: Ivo Dinov is giving an [http://portal.ncibi.org/gateway/tandtarchive.php NCIBI Tools and Technology seminar] entitled ''SOCR Infrastructure for Technology-enhanced Trans-disciplinary Health Research &amp;amp; Science Education''.&lt;br /&gt;
* Jan 14-17, 2014: [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_intro Joint Mathematics Meeting (AMS/IMS/MAA)]:&lt;br /&gt;
** January 16, 2014: Ivo Dinov is organizing a [[SOCR_Events_JMM_BigData_Jan2014 |special American Mathematical Society (AMS) session on Big Data: Mathematical and Statistical Modeling, Tools, Services, and Training]], at the [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_ss18.html2014 Joint Mathematics Meeting (JMM), Baltimore, MD, USA].&lt;br /&gt;
** January 14, 2014: Ivo Dinov (SOCR), Dennis Pearl (MBI/OSU) and Kyle Siegrist (VLPS/UAH) are organizing a continuing education workshop: [http://www.distributome.org/meetings/JMM_2014/ 2014 Distributome JMM Workshop: Interactive Probability Instruction], [http://jointmathematicsmeetings.org/meetings/national/jmm2014/2160_program_tuesday.html Stadium Ballroom 4, 2nd Floor], [http://www.marriott.com/hotels/travel/bwiih-baltimore-marriott-inner-harbor-at-camden-yards Marriott Inner Harbor].&lt;br /&gt;
&lt;br /&gt;
==2013==&lt;br /&gt;
* Dec 11 and Dec 18, 2013: Ivo Dinov is giving a [http://www.sph.umich.edu/iscr/news_events/event.cfm?ID=3323 2-part presentation about the State of the SOCR Project: Part 1: Scope, Projects and Organization; Part 2: Exploratory Data Analyses (EDA)] (Videos: [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_1/ Part 1: EDA] and [http://socr.ucla.edu/docs/videos/SOCR_UMSN_2013_2/ Part 2: Tools]).&lt;br /&gt;
* Nov 15, 2013: Ivo Dinov is a keynote speaker at [http://neuroimagen2013.eventos.cimat.mx/ 2013 CIMAT Neuro Imaging], [http://www.cimat.mx/ El Centro de Investigación en Matemáticas A.C. (CIMAT) y el Instituto de Neurobiología de la UNAM], Guanajuato, Mexico. His talk is entitled [http://neuroimagen2013.eventos.cimat.mx/node/44 Biomedical Informatics: Mathematical Techniques, Computational Challenges &amp;amp; Imaging-Genetics Applications].&lt;br /&gt;
* Oct 19, 2013: Nicolas Christou, Albert Wong and Ivo Dinov, present a [[SOCR_Events_LAUSD_Mar2013| SOCR blended approach for teaching the California Common Core State Standards for Probability and Statistics (for LAUSD Teachers)]]. &lt;br /&gt;
* Oct 18, 2013: Ivo Dinov is giving a talk entitled &amp;quot;Computational Challenges in Neuroimaging-Genetics: Predicting MCI conversion to AD&amp;quot; at the University of Michigan, Neurodegenerative Disease Research Seminar series.&lt;br /&gt;
* Oct 16, 2013: Ivo Dinov is Chairing a panel [http://michbio.org/displaycommon.cfm?an=1&amp;amp;subarticlenbr=351 discussion on &amp;quot;Big-Data&amp;quot;] at the [http://www.michbio.org 2013 Annual meeting of the Michigan Bio-Industry, MichBio]; 8:00-9:00 am on October 16, 2013, Radisson Plaza Hotel at Kalamazoo Center, Kalamazoo, MI.&lt;br /&gt;
* June 26, 2013: Ivo Dinov is speaking about &amp;quot;Analytical Pipeline Workflows, Resource Interoperability and Processing of “Big” Genomics Data&amp;quot; at a session on [http://www.clinicalgenomeconference.com/TCGC_content.aspx?id=123666&amp;amp;libID=123614 The Science of Investigation and Interpretation] at [http://www.clinicalgenomeconference.com/ The Clinical Genome Conference (TCGC)], [http://www.jdvhotels.com/hotels/california/san-francisco-hotels/hotel-kabuki Hotel Kabuki 1625 Post Street, San Francisco, CA 94115].&lt;br /&gt;
* June 16, 2013: Ivo Dinov is organizing a [[SOCR_Events_OHBM2013_Workflows|Continuing Education Course entitled Neuroimaging ‘Big Data’ Challenges and Computational Workflow Solutions]]. This workshop is part of the [http://www.humanbrainmapping.org/OHBM2013/ 2013 Organization for Human Brain Mapping Meeting] in Seattle, Washington. Ivo Dinov is presenting [http://www.humanbrainmapping.org/files/2013MeetingFiles/Sessions/Neuroimaging%20%E2%80%98Big%20Data%E2%80%99%20Challenges%20and%20Computational%20Workflow%20Solutions.pdf The Pipeline Workflow Environment] at this workshop.&lt;br /&gt;
* May 20, 2013: Ivo Dinov is organizing a [http://www.loni.ucla.edu/twiki/bin/view/LONI/PL_ComputationalGenomicsWorkshop_2013 Computational Genomics Training Workshop] at UCLA. This workshop will be postponed to a future date, due to unforeseen scheduling conflicts.&lt;br /&gt;
* May 15-16, 2013: USCOTS 2013 Workshop: Dennis Pearl (Ohio State), Kyle Siegrist (University of Alabama) and Ivo Dinov (UCLA) are organizing a 2 day workshop [http://www.distributome.org/meetings/USCOTS_2013/ Interactive Probability Instruction], at the [http://www.causeweb.org/uscots/workshops/ USCOTS 2013 meeting]. The organizers will provide continuing education training for using the [http://www.Distributome.org Probability Distributome webapps], classroom use of the [http://www.math.uah.edu/stat Virtual Laboratories in Probability and Statistics], exploratory data analysis using the [http://www.SOCR.ucla.edu Statistics Online Computational Resource].&lt;br /&gt;
* April 08, 2013: Ivo Dinov is presenting the [http://www.loni.ucla.edu/twiki/pub/LONI/EpiBioS_Bioinformatics_2013/Dinov_EpiBioS_PipelineWorkflows_2013.pptx Pipeline Workflow Environment (PPTX)] at the [http://www.loni.ucla.edu/twiki/bin/view/LONI/EpiBioS_Bioinformatics_2013 Epilepsy Bioinformatics Workshop at UCLA].&lt;br /&gt;
* March 21, 2013: Nicolas Christou and Ivo Dinov, presented the [[SOCR_Events_LAUSD_Mar2013 |SOCR blended mathematics and statistics education resources relative to the California Common Core State Standards, LAUSD]].  &lt;br /&gt;
* March 4-6, 2013: Ivo Dinov is presenting a paper entitled [http://www.pacificrimneuroimaging.net/wp-content/uploads/2012/11/Dinov_BigDataChallenges_2013.pdf How to Ride the Perfect Neuroimaging-Genetics-Computation Storm? Collision of Peta Bytes of Data, Thousands of Software Tools and Millions of Hardware Devices] at the [http://www.pacificrimneuroimaging.net/ New Horizons in Human Brain Imaging], Turtle Bay Resort, Oahu, Hawaii USA. &lt;br /&gt;
* January 23-25, 2013: [http://ccliconference.org/2013-tuesccli-pi-conference/ 2013 AAAS TUES/CCLI  conference in Washington, DC]; American Association for the Advancement of Science, AAAS, and the National Science Foundation, NSF, Programs for Course, Curriculum, and Laboratory Improvement, CCLI, and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES.&lt;br /&gt;
** Ivo Dinov is [http://ccliconference.org/abstracts/943 presenting the SOCR resource].&lt;br /&gt;
** Kyle Siegrist, Dennis Pearl and Ivo Dinov are presenting [http://ccliconference.org/abstracts/1166 the Probability Distributome Project].&lt;br /&gt;
&lt;br /&gt;
==2012==&lt;br /&gt;
* November 27, 2012: Ivo Dinov (SOCR/UCLA), Dennis Pearl (MBI/OSU) and Kyle Siegrist (UAH/VLPS) are presenting a [http://www.distributome.org/meetings/CAUSEWeb_Webinar2012/ CAUSEWeb Webinar: Hands-on Distributome Activities for Teaching Probability] (Tuesday, 2:30 PM Eastern Time).&lt;br /&gt;
* October 20-21, 2012: [http://www.google-melange.com/gsoc/homepage/google/gsoc2012 Google 2012 Summer of Code (GSoC'12)] [http://gsoc-wiki.osuosl.org/index.php/2012 Mentor Summit]: Dr. Dinov presented the [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR-GSoC'12 projects] and lead 2 working sessions on [https://gsoc-wiki.osuosl.org/index.php/Building,_supporting_and_funding_academic-industry_collaborations_-_ideas,_partners,_challenges_and_prospects Building, supporting and funding academic-industry collaborations - ideas, partners, challenges and prospects]  and [http://gsoc-wiki.osuosl.org/index.php/JS_SciCompLib JavaScript Science Computation Library] at this event.&lt;br /&gt;
* October 09, 2012: Dr. Ivo Dinov consults on a PBS NOVA/WGBH [http://www.pbs.org/wgbh/nova/body/mapping-the-brain.html Interactive Mapping the Brain Webapp] providing understanding of the relationship between brain anatomy, function and physiology. This resource enables the exploration of complex connections between different brain regions, and showcases the difference between normal and abnormal brain variability.&lt;br /&gt;
* August 02, 2012: Ivo Dinov is presenting an invited talk entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=303516 Computational Neuroinformatics: Challenges, Methods &amp;amp; Tools] at the [http://www.amstat.org/meetings/jsm/2012/ 2012 JSM meeting] in San Diego, CA. Session, (207361), [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=207361 Innovations in Biomaging and the Impact of Statistics”, Session #587], Thursday, 8/2/2012, from 8:30 AM - 10:20 AM.&lt;br /&gt;
* July 31, 2012: Scott Kamino, Bilal Bhakhrani, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2012/onlineprogram/AbstractDetails.cfm?abstractid=306019 Bivariate Normal Distribution: A New SOCR Applet] at [http://www.amstat.org/meetings/jsm/2012/onlineprogram/ActivityDetails.cfm?SessionID=208039 2012 JSM Meeting in San Diego, CA] (Tue, 7/31/2012, 10:30 AM - 12:20 PM, CC-Room 30C). &lt;br /&gt;
* July 13, 2012: Ivo Dinov is giving a lecture entitled [http://www.mntp.pitt.edu/Symposium/Symposium_DataProc_2012.html Multimodal Neuroimaging and Distributed Computing using the Pipeline Environment] at the 2012 [http://www.mntp.pitt.edu/ Multimodal Neuroimaging Training Program (MNTP)] at the University of Pittsburgh.&lt;br /&gt;
* May 2012: [http://socr.ucla.edu/htmls/ana/ SOCR Analyses] are utilized as the key statistical computing library part of the [http://pipeline.loni.ucla.edu/ Pipeline Graphical Workflow Environment]. One example is the use of the [[SOCR_EduMaterials_AnalysisActivities_MLR|SOCR Multivariate Linear Regression]] and [[SOCR_EduMaterials_AnalysesCommandLineFDR_Correction|False Discovery Rate (FDR)]] utilities for analyzing multidimensional neuroimaging data (http://ucla.in/Ie80ps).&lt;br /&gt;
* April 26, 2012: Ivo Dinov and Nicholas Christou are presenting a special session entitled [[SOCR_Events_NCMT_2012 |Technology-Enhanced Mathematics and Statistics Education]] at the [http://www.nctm.org/conferences/content.aspx?id=29461 2012 National Council of Teachers of Mathematics (NCMT) Meeting] in Philadelphia, Pennsylvania (April 25-28, 2012).&lt;br /&gt;
* April 17, 2012: Jack Van Horn, Ivo Dinov, Paul Eggert and Arthur Toga organized a workshop [http://pipeline.loni.ucla.edu/training/bigdata_2012/ Big Data Analysis Using the LONI Pipeline].&lt;br /&gt;
* March 2012: SOCR Projects are part of GSoC 2012: [http://www.google-melange.com/gsoc/org/google/gsoc2012/socr SOCR is one of the high-profile projects included in the 2012 Google Summer of Code (GSoC)]. As part of this program, SOCR is looking for highly-motivated, technologically-savvy and result-oriented students to work on [[Available_SOCR_Development_Projects|expanding the large suite of SOCR computational libraries, designing and implementing novel open-source HTML5/JavaScript/JQuery webapps and web-resources for improving science education]].&lt;br /&gt;
* March 16-17, 2012: Ivo Dinov is organizing a [http://pipeline.loni.ucla.edu/training/cchmc2012/ Pipeline Training Workshop] at the University of Cincinnati and the Cincinnati Children’s Hospital Medical Center (CCHMC), Cincinnati, OH.&lt;br /&gt;
&lt;br /&gt;
==2011==&lt;br /&gt;
* September 2011: SOCR Applications in Human Brain Mapping. [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract Human brain mapping investigators use SOCR resources and tools to investigate frontal, temporal, and ventricular dysmorphology in schizophrenia, putamen/pallidum enlargements, medication exposure in early studies, and genetic predisposition for schizophrenia]. This report published in the [http://onlinelibrary.wiley.com/doi/10.1002/hbm.21349/abstract HBM Journal] reveals the role of specific genetic or environmental risk factors toward altered brain morphology in schizophrenia.&lt;br /&gt;
* Sept. 17, 2011: Ivo Dinov is interviewed by [http://www.dailybruin.com/index.php/section/on_the_record Daily Bruin &amp;quot;On The Record&amp;quot;] about [http://www.dailybruin.com/index.php/multimedia/43623 student loans and the rate of increase of the cost of college tuition].&lt;br /&gt;
* August 12, 2011: Scott Kamino, Lance Perry and Mat Swatek submitted a [http://www.youtube-nocookie.com/embed/kVhorHV35To video entitled &amp;quot;To Stat, or not to Stat?&amp;quot;] responding to the [http://www.amstat.org/youtube/ ASA contest to promote the practice and profession of statistics].&lt;br /&gt;
* August 03, 2011: Dave Zes, Nicolas Christou and Ivo Dinov are presenting a paper entitled [http://www.amstat.org/meetings/jsm/2011/onlineprogram/AbstractDetails.cfm?abstractid=302579 Spatial Statistics and Cartography Using SOCR (Abstract #302579)] in [http://www.amstat.org/meetings/jsm/2011/onlineprogram/MainSearchResults.cfm?SponsorID=206 Session # 524 (Wed, 8/3/2011, 11:20 AM)] at the [http://www.amstat.org/meetings/jsm/2011 Joint Statistical Meeting] in Miami Beach, Florida, July 30–August 4, 2011.&lt;br /&gt;
* July 15, 2011: Dr. Dinov was interviewed and quoted in an article in the Wall Street Journal about [http://blogs.wsj.com/numbersguy/happy-square-prime-sandwich-day-1072/ the proliferation of days designated to mark certain numerical properties - Happy Square-Prime Sandwich Day]. &lt;br /&gt;
* June 24-25, 2011: Ivo Dinov is presenting [http://socr.ucla.edu/Ed/UC_OSI/ SOCR resource utilization in the University of California - Online Statistics Instruction (UC-OSI)] program. [http://www.ucop.edu UCOP] [http://groups.ischool.berkeley.edu/onlineeducation/ OIPP] [http://groups.ischool.berkeley.edu/onlineeducation/project-participants/course-design-conference Meeting, Berkeley, California]. See the video on [http://www.youtube.com/watch?v=O9aed06zCF0&amp;amp;list=PL5AB77EEBD0A8F7F9&amp;amp;index=6 YouTube].&lt;br /&gt;
* April 11-14, 2011: [http://www.healthgrid.org/news/index.php?id=31 outGRID workshop, Vilnius, Lithuania]. Ivo Dinov is presenting a paper entitled [http://www.egi.eu/indico/getFile.py/access?contribId=114&amp;amp;sessionId=12&amp;amp;resId=2&amp;amp;materialId=slides&amp;amp;confId=207 Visual Informatics and Computational Genomics using the Graphical Pipeline Environment].&lt;br /&gt;
* January 26-28, 2011: Ivo Dinov is [http://ccliconference.org/abstracts/663 presenting the SOCR resource] at the [http://www.ccliconference.com/ 2011 AAAS TUES/CCLI  conference in Washington, DC] (American Association for the Advancement of Science, AAAS, and the National Science Foundation (NSF) Programs for Course, Curriculum, and Laboratory Improvement (CCLI) and Transforming Undergraduate Education in Science, Technology, Engineering, and Mathematics, TUES).&lt;br /&gt;
* January 07, 2011:  [http://distributome.org/ Distributome-An Interactive Web-based Resource for Probability Distributions]: Kyle Siegrist, Ivo Dinov, and Dennis Pearl will present the [http://www.mathaware.org/meetings/national/jmm/2125_program_friday.html State of the Probability Distributome Project], 2011 JMM meeting, NSF Division of Undergraduate Education, Napoleon A1-A3, 3rd Floor, Sheraton, 2:00-4:00PM.&lt;br /&gt;
&lt;br /&gt;
==2010==&lt;br /&gt;
* December 04-05, 2010: [http://www.cmc-math.org/activities/north_program.html 2010 CMC-North Annual Asilomar Mathematics Conference]. Ivo Dinov will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* November 18, 2012: Dr. Dinov presented a seminar entitled [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=showevent&amp;amp;ncmd=calmonth&amp;amp;cal=cal2&amp;amp;id=1287690501-25406-1&amp;amp;ncals=&amp;amp;y=2010&amp;amp;m=11&amp;amp;d=18&amp;amp;token=&amp;amp;sb=0&amp;amp;cf=cal&amp;amp;lc=calmonth&amp;amp;swe=1&amp;amp;set=1&amp;amp;sa=0&amp;amp;sort=e,m,t&amp;amp;ws=0&amp;amp;sib=1&amp;amp;de=1&amp;amp;tf=0 Computational Neuroinformatics: Challenges, Methods, &amp;amp; Tools], UCLA [http://www.cs.ucla.edu/cgi-bin/webevent.cgi?cmd=calmonth&amp;amp;ncmd=startup&amp;amp;cal=cal2 Computer Science Seminar Series].&lt;br /&gt;
* November 05-06, 2010: [http://www.cmc-math.org/activities/south_conference.html 2010 CMC-South Annual Palm Springs Mathematics Conference]. Ivo Dinov and Nicolas Christou will [[SOCR_Events_CMC_2010 |present the SOCR Resource]].&lt;br /&gt;
* September 2010: [http://magazine.amstat.org/blog/2010/09/01/hands-on/ The SOCR MotionCharts Project receives an honorable mention in the 2010 ASA Hands-On Statistics Activity Competition]. &lt;br /&gt;
* September 2010: In a High Speed Network partnership with K-12 educators, the SOCR Probability and Statistics [[EBook]] is posted on the [http://www.k12hsn.org/calaxy/wikis.php/calaxy/Probability_and_Statistics_EBook K12HSN network and made freely and openly available to the entire K-12 educational community].&lt;br /&gt;
* August 13, 2010, [[SOCR_Events_Aug2010 | SOCR AP Statistics Continuing Education Workshop]] at UCLA.&lt;br /&gt;
* July 31 - August 05, 2010, [http://www.amstat.org/meetings/jsm/2010/ JSM 2010 Conference], Vancouver, BC, Canada, July 31 – August 5, 2010:&lt;br /&gt;
** Juana Sanchez, Nicolas Christou and Ivo Dinov will present [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR CLT Applet and Activity]], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=307049 Abstract #307049], Tue, 8/3/2010, 2:00 PM.&lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Enhancing the Teaching of Statistics:  Analysis of Spatial Data Using SOCR and R], Session #478, [http://www.amstat.org/meetings/jsm/2010/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=308613 Abstract #308613], Wed, 8/4/2010, 8:30 AM.&lt;br /&gt;
* July 20-23, 2010 [http://sloanconsortium.org/et4online Sloan Consortium/MERLOT symposium] (Fairmont Hotel in San Jose, CA). Nicolas Christou, Ryan Rosario and Ivo Dinov will present a session entitled [[SOCR_Events_SloanMerlot_2010 | Technology-Enhanced Probability and Statistics Education Using the Statistics Online Computational Resource]].&lt;br /&gt;
* June 09, 2010: SOCR V2.6 receives the [http://www.softpedia.com/progClean/Statistics-Online-Computational-Resource-Clean-161967.html Softpedia &amp;quot;100% Free &amp;amp; Clean&amp;quot; award]. This [http://www.socr.ucla.edu/htmls/SOCR_Recognitions.html certifies that all SOCR tools] are free of all forms of malware, including spyware, viruses, trojans and backdoors, and can be used and installed with no concern by any computer user.&lt;br /&gt;
* May 14, 2010: Dr. Dinov was interviewed and quoted in an article in the [http://blogs.wsj.com/numbersguy/overblown-notions-about-names-940/ Wall Street Journal about the concept of implicit egotism, or people’s unconscious attraction to things or people that resemble themselves — as manifested by name similarity].&lt;br /&gt;
* May 11, 2010: [[SOCR_Events_CAUSE_TL_Spring2010 | CAUSE SOCR Teaching and Learning Webinar]].&lt;br /&gt;
* April 19, 2010: [[SOCR_Events_FTO_Obesity_Neuroimaging2010 |The largest neuroimaging study to-date identifies the relations between obesity gene (FTO) and brain tissue loss using the SOCR computational libraries]].&lt;br /&gt;
* March 25, 2010: We upgraded the [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR user web-statistics mechanism]. Now we provide anonymous access to dynamic and complete statistics on the web-users of the main [http://www.SOCR.ucla.edu SOCR server]. The new [http://www.socr.ucla.edu/SOCR_UserGoogleMap.html SOCR web-statistics architecture] allows anyone to see anytime the locations and types of the last 500 SOCR users. Each user is uniquely counter daily, irrespective of usage-time or number of SOCR resources they used. Other [[SOCR_Servers | SOCR servers]] have separate tracking mechanisms.&lt;br /&gt;
* March 17, 2010: [http://www.scientistlive.com/European-Science-News/Medical/Brain_abnormalities_in_children_exposed_to_meth/24237/ Brain abnormalities in children exposed to meth] - a news-brief on a recent [http://www.jneurosci.org/cgi/content/abstract/30/11/3876 article on Prenatal Methamphetamine use], co-authored by Dr. Dinov, is distributed by the National Institutes of Health.&lt;br /&gt;
* January 12, 2010: NSF/DUE PI Meeting at the [http://www.ams.org/amsmtgs/2124_intro.html 2010 JMM Meeting in San Francisco, CA]. Dr. Dinov [[SOCR_Events_JMM_Jan2010 |presented the SOCR Resource]].&lt;br /&gt;
&lt;br /&gt;
==2009==&lt;br /&gt;
* October 2009: A new major upgrade of [[SOCR_EduMaterials_AnalysesActivities |SOCR Analyses]] is released that fixed the appearance of the tabs on all SOCR analysis applets. The new version only shows the tabs that are appropriate for the specific SOCR Analysis chosen by the user. This upgrade of Analyses resolves this inconsistency and makes the navigation to data-input, variable-mapping and result-interpretation much more intuitive and user-friendly.&lt;br /&gt;
* September 2009: A new [http://forums.stat.ucla.edu/socr SOCR forum is introduced]. Users, instructors, students and developers may use this forum to post questions, learn about various SOCR datasets, tools, learning materials, and contribute to the SOCR knowledgebase.&lt;br /&gt;
* August 15-23, 2009 [http://www.statssa.gov.za/isi2009/index.aspx SOCR Organized Events at the 2009 International Statistical Institute Meeting]&lt;br /&gt;
** August 16, 2009, [[SOCR_Events_ISIW_Aug2009 | SOCR Workshop at 2009 ISI Meeting]].&lt;br /&gt;
** August 17, 2009, [[SOCR_Events_ISIS_Aug2009 | A special session entitled ''Interactive, Data-Driven and Technology-Enhanced Approaches for Probability and Statistics Education'']], organized by Ivo Dinov and Nicolas Christou at 2009 ISI Meeting.&lt;br /&gt;
* August 10, 2009: The [http://www.dailybruin.com/index.php/article/2009/08/ucla-professors-host-workshop-open-source-projects Daily Bruin published an article about SOCR], summarizing [http://wiki.stat.ucla.edu/socr/uploads/9/93/DailyBruin_UCLA_SOCR_Aug2009.pdf  the project scope and interviewing Ivo Dinov and Nicolas Christou].&lt;br /&gt;
* August 10-12, 2009, [[SOCR_Events_Aug2009 | SOCR Training &amp;amp; Development Workshop]] at UCLA.&lt;br /&gt;
* August 03, 2009, [http://www.amstat.org/meetings/jsm/2009/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=164&amp;amp;sessionid=204418 Confidence Intervals using SOCR]. Nicolas Christou and Ivo Dinov are presenting SOCR at the [http://www.amstat.org/meetings/jsm/2009/ 2009 JSM Meeting in Washington, DC].&lt;br /&gt;
* June 27, 2009, Nicolas Christou [http://www.causeweb.org/uscots/uscots09/program/posters.pdf presented a poster (page 19)] at the [http://www.causeweb.org/uscots/uscots09/ 2009 USCOTS conference at OSU].&lt;br /&gt;
* April 16, 2009, SOCR Sponsored [http://cts.stat.ucla.edu/seminars/ Teaching Statistics Seminar] by [http://www.stat.tamu.edu/~west/ Webster West], 4-5 PM, Math Science 5137.&lt;br /&gt;
* January 05-08, 2009: Ivo Dinov will give 2 talks and present a poster at [http://www.ams.org/amsmtgs/2110_program.html 2009 Joint AMS/MAA Meeting in Washington, DC], January 05-08, 2009.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf Interactive, Data-Driven and Technology-Enhanced Approach for Probability and Statistics Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:MAACPE1  Demos and Strategies with Technology that Enhance Teaching and Learning Mathematics], Tuesday, January 06, 2009, 9:20AM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-e1-26.pdf 1046-E1-26], [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/SOCR_AMS_1046_E1_26_01_06_09.pdf PDF Slides].&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS Statistics Online Computational Resource for Education], at a session on [http://www.ams.org/amsmtgs/2110_program_tuesday.html#2110:SCOTTPS  MAA Poster Session on Projects Supported by the NSF Division of Undergraduate Education], Tuesday, January 06, 2009, 2 PM, Blue Room and Foyer, Omni Shoreham Hotel.&lt;br /&gt;
** [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf Integrated, Multidisciplinary and Technology-Enhanced Science Education: The Next Frontier], at a [http://www.ams.org/amsmtgs/2110_program_wednesday.html#2110:AMSCP28 Mathematics Education session], Wednesday, January 07, 2009, 1:15PM, [http://www.ams.org/amsmtgs/2110_abstracts/1046-97-25.pdf 1046-97-25], [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.ppt PPT Slides] and  [http://www.socr.ucla.edu/docs/NISER_AMS_1046_97_25_01_07_09.pdf PDF Slides].&lt;br /&gt;
&lt;br /&gt;
== 2008 ==&lt;br /&gt;
* November 17, 2008: [[SOCR_Overview_2008 | SOCR Presentation]] at [http://courses.stat.ucla.edu/index.php?term=08F&amp;amp;course=6637912&amp;amp;lecture=66379120 Statistics 495A] (Fall 2008).&lt;br /&gt;
* October 17, 2008: [[SOCR_Events_UCCP2008 | SOCR presentation at the UCCP/UCLA CDI meeting]] at at [http://www.cdi.ucla.edu/ UCLA CDI].&lt;br /&gt;
* September 27, 2008: Ivo Dinov is [[SOCR_Events_SMC2008 | presenting the SOCR resources]] at a continuing education event at Santa Monica College.&lt;br /&gt;
* September 17, 2008: The complete [[SOCR]] source code, v.2.5, was publicly released via the [http://socr.googlecode.com SOCR Google Code project page]. This new release contains the latest version of the [[SOCR]] libraries and all additional resources needed to revise, compile, package and deploy the entire [[SOCR]] resource on the Internet.&lt;br /&gt;
* August 14-15, 2008: Ivo Dinov is [[SOCR_Events_CCLI2008 | presenting the SOCR resource]] at the [http://www.ccliconference.com/ 2008 CCLI AAAS conference in Washington, DC].&lt;br /&gt;
* August 9, 2008: Nicolas Christou will present a seminar entitled [http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09#1090 Law of Large Numbers: The Theory, Applications and Technology-based Education], at the [http://mic08.merlot.org/program 2008 MERLOT International Conference] ([http://mic08.merlot.org/program/index.php?sort=date&amp;amp;date=2008-08-09 1:30-2:00 PM in the Conrad A Room]).&lt;br /&gt;
* August 2008: SOCR is releasing a new [http://www.esurveyspro.com/Survey.aspx?id=79ba4c38-b7d0-4530-aa00-12fdd32b6609 electronic polling survey to solicit open and anonymous community feedback on existing SOCR resources and recommendations for development of future SOCR materials and tools].&lt;br /&gt;
* August 2008, [http://www.amstat.org/meetings/jsm/2008 2009 Joint Statistics Meeting]&lt;br /&gt;
** Thu, 8/7/08, 10:30 AM - 12:20, Nicolas Christou is Chairing a session entitled [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=519&amp;amp;sessionid=203496 The Internet, Telecommunications, and Migration]&lt;br /&gt;
** Wed, 8/06/08, 2:00 PM - 3:50 PM, Nicolas Christou and Ivo Dinov are presenting [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=abstract_details&amp;amp;abstractid=301649 Applications of Statistics in Finance Using the Statistics Online Computational Resource (SOCR)] in a special session on [http://www.amstat.org/meetings/jsm/2008/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=451&amp;amp;sessionid=203500 Portfolio Analysis, Exchange Rates, Microstructure, and GARCH Models] &lt;br /&gt;
* July 29, 2008: Ivo Dinov will give a hands-on demonstration entitled [[SOCR_Events_July2008 | Using SOCR to Motivate Simulation Experiments in Middle and High School]] at the [http://censusatschool-california.stat.ucla.edu/ Second CensusAtSchool International Workshop], [http://censusatschool-california.stat.ucla.edu/ 3:15-4:00 PM, CCLIC Lab].&lt;br /&gt;
* July 07, 2008: [http://www.math.ucla.edu/~lvese Luminita Vese] and [http://www.stat.ucla.edu/~dinov Ivo Dinov] are co-organizing a [http://www.siam.org/meetings/is08/ SIAM Imaging Science] session on [http://meetings.siam.org/program.cfm?CONFCODE=IS08 Computational Science and Biology: The Challenges, Data, Methods and Tools] (MS4), [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6978 Part I] and [http://meetings.siam.org/sess/dsp_programsess.cfm?SESSIONCODE=6979 Part II].&lt;br /&gt;
* June 20, 2008: Nicolas Christou will present a seminar entitled [http://trc.ucdavis.edu/uc21st/concurrent.html Web-based, Data-driven Probability and Statistics Education], at the [http://trc.ucdavis.edu/uc21st/ UC21st Century - Teaching, Learning and Technology: Past, Present and Future] Conference ([http://trc.ucdavis.edu/uc21st/schedule.html 11:45AM - 12:30PM]). Video ([http://uocatdavis.wmod.llnwd.net/a2636/e1/TLTC/TLTC_Nicholas_Christou.wmv WMP], [http://webcast.ucdavis.edu/TLTC/ FLASH]) and [http://trc.ucdavis.edu/uc21st/proceedings/christou_dinov_uc21st.ppt PPT slides] ara available online.&lt;br /&gt;
* June 05, 2008: [[SOCR_Events_EOY2008 | SOCR End-of-Year Event]].&lt;br /&gt;
* [http://cts.stat.ucla.edu/seminars 2008 SOCR/CTS '''Seminars''']:&lt;br /&gt;
** Thursday May 29: [http://www.stat.berkeley.edu/~stark/ Phil Stark], [http://www.stat.berkeley.edu/ Department of Statistics, UC Berkeley], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, May 15: [http://www.stat.ucla.edu/~esfandia Mahtash Esfandiari] and Hai Nguyen,  Dept. of Statisitics, UCLA&lt;br /&gt;
** Thursday, May 1: [http://www.stat.osu.edu/~dkp/ Dennis Pearl], [http://www.stat.osu.edu/ Dept. of Statistics, The Ohio State University]. &amp;quot;The Statistical Buffet&amp;quot; course redesign at OSU, [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday, April 24: [http://mathequity.terc.edu/gw/html/Andeepage.html Andee Rubin], [http://www.terc.edu/ TERC], [http://cts.stat.ucla.edu/seminars/rubin.html Software as a Learning Context: the Case of TinkerPlots and Statistical Reasoning], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
** Thursday April 10:  [http://www.maths.qut.edu.au/profiles/macgillivray/ Helen MacGillvray], Queensland University of Technology, Australia, [http://cts.stat.ucla.edu/seminars/macgillvray.html Roles of Assessment in Learning of Statistics and Mathematics], [http://cts.stat.ucla.edu/seminars/ Thursday, 4-5pm, Math Science # 5137].&lt;br /&gt;
* May 19, 2008: Dr. [http://www.stat.ucla.edu/~nchristo Nicolas Christou] will be awarded the prestigious [http://www.oid.ucla.edu/edtech/bpcaward Brian P. Copenhaver Award for Innovation in Teaching with Technology]. [[SOCR_Awards_Christou_BC2008 | More information about this event and Nicolas' contributions is available here]]. Nicolas Christou [http://www.oid.ucla.edu/edtech/bpcaward/2008-video-christou (video-stream) Interview] and [http://www.today.ucla.edu/campus/080520_online_stats-aid/ UCLA Today] article.&lt;br /&gt;
* May 05, 2008: Ivo Dinov will give a [http://www.amstat.org ASA]/[http://www.amstat.org/education/gaise/ GAISE] [[SOCR_Events_May2008 | Webinar for middle school teachers in various science and quantitative disciplines]].&lt;br /&gt;
* January 17-19, 2008, [http://www.hicstatistics.org/program_stats.htm 2008 Hawaii International Conference on Statistics, Mathematics and Related Fields], Honolulu, Hawaii: &lt;br /&gt;
** Nicolas Christou and Ivo Dinov will present [http://www.hicstatistics.org/STATS2008.pdf Statistics Online Computational Resource for Education: www.socr.ucla.edu];&lt;br /&gt;
** Priscilla Chui, Nicolas Christou and Ivo Dinov will present a talk on [http://www.hicstatistics.org/STATS2008.pdf Numbers and Sense].&lt;br /&gt;
* January 6-9, 2008: [http://www.ams.org/amsmtgs/2109_program.html Joint Mathematics Meetings (AMS/MAA), San Diego, CA]. &lt;br /&gt;
** Annie Che, Nicolas Christou, Jenny Cui and Ivo Dinov will present a poster entitled [http://www.ams.org/amsmtgs/2109_program_monday.html SOCR Analyses: a free Internet-based Statistical Analysis Toolkit], on 01/07/08, 2-4 PM, as part of the [http://www.ams.org/amsmtgs/2109_program_monday.html NSF Division of Undergraduate Education] session.&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-P1-707) [http://www.ams.org/amsmtgs/2109_program_wednesday.html#2109:MAACPP2 Interactive Web-Based Probability Distribution Mathlets: www.SOCR.ucla.edu] at the [http://www.ams.org/amsmtgs/2109_abstracts/1035-p1-707.pdf MAA Session Mathlets and Web Resources for Mathematics and Statistics Education], [http://www.ams.org/amsmtgs/2109_program_wednesday.html  1:20 PM, Wednesday Jan. 9, 2008].&lt;br /&gt;
** Rahul Gidwani, Nicolas Christou and Ivo Dinov will present a paper (1035-62-1789) [http://www.ams.org/amsmtgs/2109_abstracts/1035-62-1789.pdf Generating Functions: Web-based SOCR Applets and Computational Library Interfaces], [http://www.ams.org/amsmtgs/2109_program_wednesday.html 3:45 PM on Jan 09, 2008].&lt;br /&gt;
&lt;br /&gt;
== 2007 ==&lt;br /&gt;
* Dec. 19. 2007, SOCR was added as a [http://apps.facebook.com/socrapp/ Facebook Application] and made available as a tool/plug-in to all FaceBook Users. To add the SOCR App to your Facebook page follow these steps:&lt;br /&gt;
** [http://www.facebook.com Login your Facebook account (or create one)]&lt;br /&gt;
** [http://apps.facebook.com/socrapp Visit the SOCR App home page]&lt;br /&gt;
** [http://www.facebook.com/apps/application.php?api_key=f7ba27416638bfa874514da430932a04 Add the SOCR App to your Applications by selecting '''add this application button''']&lt;br /&gt;
* Dec. 07, 2007, Dr. Dinov was interviewed and quoted in [http://wiki.stat.ucla.edu/socr/uploads/f/f9/WSJ_12_07_2007.pdf An article] in the [http://online.wsj.com/article/SB119698695198016514.html Wall Street Journal] about a paper by Leif Nelson (NYU) and Joseph Simmons (Yale) entitled [http://papers.ssrn.com/sol3/papers.cfm?abstract_id=946249 Moniker Maladies: When Names Sabotage Success]. In this article the authors present experimental designs and data to support hypotheses that ''unconscious behavior can insidiously undermine conscious pursuits''.&lt;br /&gt;
* Sept. 24, 2007: The [http://www.NSF.gov National Science Foundation (NSF)] extended the funding for the UCLA [[SOCR | Statistics Online Computational Resource (SOCR)]] for another 4 years (2007-2011). This is a great accomplishment for the [http://socr.ucla.edu/htmls/SOCR_Team.html SOCR team] and recognition of the SOCR Resource achievements since 2002. In the next 4 years, SOCR will design, test, validate, and disseminate 1) tools (applets, demos, GUI interfaces), 2) educational materials (activities, class notes, tutorials), and 3) resources (Statistics Online Computational Resource Wiki, consulting, workshops, etc.) More information is available [http://www.nsf.gov/awardsearch/showAward.do?AwardNumber=0716055 online].&lt;br /&gt;
* Aug. 22-29, 2007: SOCR [http://wiki.stat.ucla.edu/socr/uploads/c/c4/SOCR_Abstract_Lisbon_ISI2007.pdf sampling and simulation resources] are going to be presented as an invited talk (Nicolas Christou) at the 2007 [http://www.isi2007.com.pt International Statistics Institute meeting (ISI)], Lisbon, Portugal, ([http://wiki.stat.ucla.edu/socr/uploads/7/7b/Christou_Dinov_Sanchez_ISI2007.pdf PDF Slides])&lt;br /&gt;
* Aug. 08-10, 2007: Nicolas Christou is invited to present the [[SOCR_EduMaterials_Activities_GeneralCentralLimitTheorem | SOCR Central Limit Theorem Applet and hands-on activity]] at the [http://mic07.merlot.org/program/index.php?sort=date&amp;amp;date=2007-08-09#904 2007 Multimedia Educational Resource for Learning and Online Teaching Conference], [http://conference.merlot.org/2007/ MERLOT], New Orleans, LA. ([http://wiki.stat.ucla.edu/socr/uploads/2/21/Christou_MERLOT_CLT_2007.pdf Talk PDF], [http://conference.merlot.org/2007/Thursday/merlot_clt_2007.ppt Talk PPT])&lt;br /&gt;
* Aug. 06-08, 2007: [[SOCR_Events_Aug2007 | SOCR/CAUSE Workshop]] at UCLA&lt;br /&gt;
* Aug. 01, 2007: [http://www.amstat.org/meetings/jsm/2007/ JSM] - Nicolas Christou will present [http://www.amstat.org/meetings/jsm/2007/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=465&amp;amp;sessionid=202134 Enhancing the Teaching of Statistics with Technology Using the Statistics Online Computational Resource (SOCR)] in Session 466, at 2:00 PM at the Joint Statistical Meetings in Salt Lake City, Utah, July 29-August 2, 2007 ([http://wiki.stat.ucla.edu/socr/uploads/5/52/Christou_Talk_JSM_Aug2007.pdf Abstract #309845], PDF).&lt;br /&gt;
* June 01 and June 08, 2007 (4PM): [http://www.socr.ucla.edu/docs/SOCR_talks_060107.html The State of the Statistics Online Computational Resources], Ivo Dinov will present at the [http://cts.stat.ucla.edu/seminars/ UCLA Statistics Teaching Seminar].&lt;br /&gt;
* May 21-22, 2007, [http://pages.stern.nyu.edu/~gsimon/ Gary Simon], [http://pages.stern.nyu.edu/ NYU Stern], will visit SOCR and give a [http://seminars.stat.ucla.edu/ seminar talk] (TBD).&lt;br /&gt;
* May 17-19, 2007: SOCR will be presented at an [http://wiki.stat.ucla.edu/socr/index.php/USCOTS_2007_Program_SOCR invited breakout session] of the [http://www.causeweb.org/uscots USCOTS 2007] (Juana Sanchez, Nicolas Christou and Ivo Dinov).&lt;br /&gt;
* Mar. 07, 2007. [http://wiki.stat.ucla.edu/socr/uploads/3/3c/TheNumbersGuy_WSJ_03_07_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117338464249431351-0ghR_0Jef5ubo6ZLbYIVoePRxrA_20070408.html?mod=tff_main_tff_top Wall Street Journal] discussing probability estimations quotes Ivo Dinov.&lt;br /&gt;
* Feb. 27, 2007. [http://wiki.stat.ucla.edu/socr/uploads/8/88/TheNumbersGuy_WSJ_02_23_2007.pdf An article] in the [http://online.wsj.com/public/article/SB117207018272214963-M0nkHhnjRCR5_SIvOWHlNw2sCeE_20070324.html?mod=tff_main_tff_top Wall Street Journal] quotes the SOCR PI, Dr. Dinov.&lt;br /&gt;
* Jan. 08, 2007. [[SOCR]] was included in the [http://www.rsscse.org.uk/news/rssnewseducation.asp Royal Statistical Society News Brief]&lt;br /&gt;
* Jan. 05-08, 2007: [[SOCR]] Resource is presented at the [http://www.ams.org/amsmtgs/2098_intro.html Joint AMS/MAA meeting] in New Orleans, LA.&lt;br /&gt;
** Annie Che will present the [http://www.ams.org/amsmtgs/2098_progfull.html SOCR Analyses], Saturday January 6, 2007, 2:00 PM&lt;br /&gt;
** Ivo Dinov will give a talk on [http://www.ams.org/amsmtgs/2098_abstracts/1023-62-1.pdf SOCR Utilization and Assessment], Monday January 8, 2007, 2:45 PM, [http://www.ams.org/amsmtgs/2098_program_monday.html#2098:AMSCP28 AMS Session on Probability and Statistics]&lt;br /&gt;
&lt;br /&gt;
== 2006 ==&lt;br /&gt;
* Ivo Dinov presented two talks at the 2006 American Medical Informatics Association ([http://www.amia.org AMIA]) [http://www.amia.org/meetings/f06/ Annual Meeting, Washington, DC]&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showdoc.asp?DID=401 Biomedical Imaging Ontologies: Design Principles Enabling Interoperability for Imaging Applications, Tools, and Data], Nov. 11, 2006&lt;br /&gt;
** [http://www.amia.org/meetings/f06/showeve.asp?EID=S14 Seeing Biomedicine: New Opportunities and Research Challenges at the Intersection of Biomedical Imaging and Informatics], Nov. 13, 2006&lt;br /&gt;
* Maria Pilar Munoz, Dept. of Statistics, North Carolina State University, visited SOCR and gave a talk on [http://seminars.stat.ucla.edu/ ''Formal Assessment of a Web-based Tool Designed to Improve Student Performance in Statistics''], Tuesday, November 21, 2006, 3-4 pm; 6627 Math Sciences Bldg. Seminar#: 297.&lt;br /&gt;
* Juana Sanchez presented [http://www.socr.ucla.edu/docs/Sanchez_SOCR_JSM_2006.pdf SOCR Usage and Validation] at the [http://www.amstat.org/meetings/jsm/2006/onlineprogram/index.cfm?fuseaction=activity_details&amp;amp;activityid=148&amp;amp;sessionid=201662 Section on Statistical Education] of the [http://www.amstat.org/meetings/jsm/2006 Joint Statistics Meeting], August 2006, Seattle, WA.&lt;br /&gt;
* Juana Sanchez presented various SOCR Activities at the [http://www.maths.otago.ac.nz/icots7 ICOTS7], July 2006.&lt;br /&gt;
* Nicolas Christou gave a SOCR Colloquium Talk at the department of [http://www.mas.ucy.ac.cy/ Mathematics &amp;amp; Statistics, University of Cyprus], June 2006.&lt;br /&gt;
* [http://www.loni.ucla.edu/CCB/training.aspx?id=2912 2006 SIAM Symposia on Brain Segmentation], May 15-17, 2006 in Minneapolis, Minnesota, USA. Ivo Dinov organized  a symposium at the [http://www.siam.org/meetings/is06 2006 SIAM Conference on Imaging Science] entitled ''Mathematical Methods and Tools for Volumetric Brain Segmentation''.&lt;br /&gt;
* Ivo Dinov presented the SOCR Resource at the [http://www.ams.org/amsmtgs/2095_intro.html Joint AMS/MAA joint Meeting], San Antonio, TX, January 2006&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: http://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php/SOCR_News}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_FAMU_Talk_2026&amp;diff=18359</id>
		<title>SOCR FAMU Talk 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_FAMU_Talk_2026&amp;diff=18359"/>
		<updated>2026-03-16T23:24:48Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Talk */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: Dinov's 2026 Talk at [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications Seminar Series] ==&lt;br /&gt;
&lt;br /&gt;
[[Image:BigData_AMS_JMM_2014.gif|250px|thumbnail|right| [https://aimlfacultycluster.lovable.app/ FAMU Data to Discovery: Tools, Trends, and Applications Seminar Series] ]]&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
==Talk==&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
* [https://aimlfacultycluster.lovable.app/ '''FAMU Data to Discovery: Tools, Trends, and Applications Seminar Series''']&lt;br /&gt;
&lt;br /&gt;
* ''Date'': Monday, March 23, 2026, 4:00 PM ET&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://nellymateeva.wixsite.com/nmateeva Nelly Mateeva-Tarkalanova (FAMU)]&lt;br /&gt;
&lt;br /&gt;
* ''Presenter'': Ivo D. Dinov (UMich)&lt;br /&gt;
&lt;br /&gt;
* '''Title''': ''The Realities of Augmented Intelligence''&lt;br /&gt;
&lt;br /&gt;
* ''Abstract'': &lt;br /&gt;
&lt;br /&gt;
: This decade, the evolution of computational intelligence (algebraic tought) has reached a critical juncture, shifting the paradigm from Artificial Intelligence (AI) as an autonomous agent to Augmented Intelligence (AI) as a collaborative partner. This talk navigates the technical and philosophical trajectory of this transition, beginning with the historical foundations of Artificial Neural Networks (ANN). In the present landscape, AI systems exhibit strong generative capabilities, they often operate as &amp;quot;black boxes&amp;quot; limited by the boundaries of classical spacetime data. We address the limitations of today’s deep learning, specifically issues of interpretability and temporal reasoning, and explore a future defined by Augmented Intelligence where Science and Arts are tightly coupled. This next phase suggests human-centric collaboration, where technology enhances rather than replaces human heuristic judgment.&lt;br /&gt;
&lt;br /&gt;
: A central technical highlight of this session is the application of Spacekime Analytics. By extending the traditional 4D Minkowski spacetime into a 5D &amp;quot;spacekime&amp;quot; manifold using complex-time (kime), we demonstrate how augmented systems can better model high-dimensional, longitudinal data. Spacekime Analytics offers a robust framework for improving inference in fields ranging from neuroimaging to global economic forecasting, providing a mathematical bridge between raw data and human-interpretable insights. We will examine the multidisciplinary implications of AI. From an academic standpoint, we discuss the &amp;quot;human premium&amp;quot; in research; on a societal level, we address the ethics of &amp;quot;agentic&amp;quot; workflows; and globally, we analyze the tension between technological acceleration and human values. The reality of augmented intelligence lies not in the creation of a &amp;quot;digital god,&amp;quot; but in the sophisticated, mathematically grounded extension of the human mind.&lt;br /&gt;
&lt;br /&gt;
* ''Slides'': [https://wiki.socr.umich.edu/images/f/f7/Dinov_AI_Reality_Slidedeck_2026.pdf Slidedeck]&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_FAMU_Talk_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=File:Dinov_AI_Reality_Slidedeck_2026.pdf&amp;diff=18358</id>
		<title>File:Dinov AI Reality Slidedeck 2026.pdf</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=File:Dinov_AI_Reality_Slidedeck_2026.pdf&amp;diff=18358"/>
		<updated>2026-03-16T23:23:55Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18357</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18357"/>
		<updated>2026-03-16T03:01:45Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Other Sessions */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
: A ''heuristic overview'' of complex-time representation, kime-phase tomography, and spacekime analytics.&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/Spacekime_TCIU_Podcast1_2025.mp3 A brief, 7-minute, spacekime ''podcast'' (audio)].&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/ComplexTimeReps_KPT_Spacekime_FieldEquations_video.mp4 A short, 5-minute ''video'' describing complex-time representation, kime-phase tomography (KPT), and spacekime gravitational field equations].&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data.&lt;br /&gt;
:: [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Kime-Phase Tomography (KPT) (slidedeck)], &lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves],&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
: [https://tciu.predictive.space/ Additional theoretical developments, experimental demonstrations, computational algorithms, and data science applications are available on the '''spacekime/TCIU website'''].&lt;br /&gt;
&lt;br /&gt;
: [https://socr.umich.edu/DSPA2/DSPA2_notes/01_Introduction.html#2_Foundations_of_R More information about editing, interpreting, knitting/compiling and running ''R''-markdown electronic notebooks (*.Rmd) is available on the SOCR DSPA2 site].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* GDS GPS general meeting: Tue, March 17, 6:45 p.m. Convention Center, Meeting Room 507, EVT-AA58, GDS Unit Business Meeting&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18356</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18356"/>
		<updated>2026-03-16T03:01:23Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Other Sessions */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
: A ''heuristic overview'' of complex-time representation, kime-phase tomography, and spacekime analytics.&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/Spacekime_TCIU_Podcast1_2025.mp3 A brief, 7-minute, spacekime ''podcast'' (audio)].&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/ComplexTimeReps_KPT_Spacekime_FieldEquations_video.mp4 A short, 5-minute ''video'' describing complex-time representation, kime-phase tomography (KPT), and spacekime gravitational field equations].&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data.&lt;br /&gt;
:: [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Kime-Phase Tomography (KPT) (slidedeck)], &lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves],&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
: [https://tciu.predictive.space/ Additional theoretical developments, experimental demonstrations, computational algorithms, and data science applications are available on the '''spacekime/TCIU website'''].&lt;br /&gt;
&lt;br /&gt;
: [https://socr.umich.edu/DSPA2/DSPA2_notes/01_Introduction.html#2_Foundations_of_R More information about editing, interpreting, knitting/compiling and running ''R''-markdown electronic notebooks (*.Rmd) is available on the SOCR DSPA2 site].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* GDS GPS general meeting: Tue, March 17, 6:45 p.m. Convention Center, Meeting Room 507 &lt;br /&gt;
EVT-AA58, GDS Unit Business Meeting&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18355</id>
		<title>SMHS GLM</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18355"/>
		<updated>2026-03-13T21:09:50Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Gamma GLM vs. Beta Regression */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Generalized Linear Modeling (GLM) ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Generalized Linear Modeling (GLM) is a flexible generalization of ordinary linear regression that allows response variables to have error distribution models other than a normal distribution. GLM extends linear regression by allowing the linear model to be related to the response variable via a link function and enabling the variance of each measurement to be a function of its predicted value. This framework unifies statistical models including linear regression, logistic regression, and Poisson regression. Estimation methods include iteratively reweighted least squares for maximum likelihood estimation, Bayesian approaches, and least squares fitted to variance-stabilized responses.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
While linear regression models linear relationships between response and predictors, many real-world scenarios involve response variables that don't follow normal distributions. For example:&lt;br /&gt;
* Binary outcomes (yes/no decisions) with probabilities bounded between 0 and 1&lt;br /&gt;
* Count data (number of events) that follow Poisson distributions&lt;br /&gt;
* Survival times that follow exponential or Weibull distributions&lt;br /&gt;
&lt;br /&gt;
GLM provides a unified framework for these situations by allowing response variables from the exponential family of distributions.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====1) GLM Components====&lt;br /&gt;
A GLM consists of three components:&lt;br /&gt;
&lt;br /&gt;
1. Random Component: The response variable \(Y\) follows a distribution from the exponential family:&lt;br /&gt;
   &amp;lt;math&amp;gt;f_Y(y|\theta,\phi) = \exp\left\{\frac{y\theta - b(\theta)}{a(\phi)} + c(y,\phi)\right\}&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;\theta&amp;lt;/math&amp;gt; is the natural parameter and \(\phi\) is the dispersion parameter.&lt;br /&gt;
&lt;br /&gt;
2. Systematic Component: The linear predictor \(\eta\):&lt;br /&gt;
   &amp;lt;math&amp;gt;\eta = \mathbf{X}\beta = \beta_0 + \beta_1X_1 + \beta_2X_2 + \cdots + \beta_pX_p&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
3. Link Function: \(g(\cdot)\) that relates the mean \(\mu = E[Y]\) to the linear predictor:&lt;br /&gt;
   &amp;lt;math&amp;gt;g(\mu) = \eta \quad \text{or equivalently} \quad \mu = g^{-1}(\eta)&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The variance function relates the variance to the mean: &amp;lt;math&amp;gt;\text{Var}(Y) = a(\phi)V(\mu)&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;V(\mu)&amp;lt;/math&amp;gt; is the variance function specific to the distribution.&lt;br /&gt;
&lt;br /&gt;
====2) Exponential Family Distributions====&lt;br /&gt;
The exponential family includes many common distributions:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Distribution !! Support !! Natural Parameter \(\theta\) !! \(b(\theta)\) !! Canonical Link \(g(\mu)\) !! Variance Function \(V(\mu)\)&lt;br /&gt;
|-&lt;br /&gt;
| Normal || \(\mathbb{R}\) || \(\mu\) || \(\frac{\theta^2}{2}\) || Identity: \(\mu\) || 1&lt;br /&gt;
|-&lt;br /&gt;
| Binomial || \(\{0,1,\ldots,n\}\) || \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(n\log(1+e^\theta)\) || Logit: \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(\mu\left(1-\frac{\mu}{n}\right)\)&lt;br /&gt;
|-&lt;br /&gt;
| Poisson || \(\mathbb{N}_0\) || \(\log(\mu)\) || \(e^\theta\) || Log: \(\log(\mu)\) || \(\mu\)&lt;br /&gt;
|-&lt;br /&gt;
| Gamma || \(\mathbb{R}^+\) || \(-\frac{1}{\mu}\) || \(-\log(-\theta)\) || Inverse: \(\frac{1}{\mu}\) || \(\mu^2\)&lt;br /&gt;
|-&lt;br /&gt;
| Inverse Gaussian || \(\mathbb{R}^+\) || \(-\frac{1}{2\mu^2}\) || \(-\sqrt{-2\theta}\) || Inverse squared: \(\frac{1}{\mu^2}\) || \(\mu^3\)&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====3) Maximum Likelihood Estimation====&lt;br /&gt;
For a GLM with \(n\) independent observations, the log-likelihood is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\ell(\beta) = \sum_{i=1}^n \frac{y_i\theta_i - b(\theta_i)}{a(\phi)} + c(y_i,\phi)&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\theta_i = \theta(\mu_i)\) and \(\mu_i = g^{-1}(\mathbf{x}_i^\top\beta)\).&lt;br /&gt;
&lt;br /&gt;
The score equations are:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathbf{U}(\beta) = \frac{\partial\ell}{\partial\beta} = \mathbf{X}^\top\mathbf{W}(\mathbf{y} - \boldsymbol{\mu}) = 0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\mathbf{W} = \text{diag}\left\{\frac{1}{a(\phi)V(\mu_i)[g'(\mu_i)]^2}\right\}\).&lt;br /&gt;
&lt;br /&gt;
The Fisher information matrix is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathcal{I}(\beta) = E\left[-\frac{\partial^2\ell}{\partial\beta\partial\beta^\top}\right] = \mathbf{X}^\top\mathbf{W}\mathbf{X}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Parameters are estimated via Iteratively Reweighted Least Squares (IRLS):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\beta^{(t+1)} = \beta^{(t)} + (\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{z}^{(t)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(z_i^{(t)} = \eta_i^{(t)} + (y_i - \mu_i^{(t)})g'(\mu_i^{(t)})\).&lt;br /&gt;
&lt;br /&gt;
====4) Deviance and Goodness-of-Fit====&lt;br /&gt;
The deviance measures goodness-of-fit:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D = 2[\ell(\text{saturated model}) - \ell(\text{fitted model})]&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
For nested models \(M_0 \subset M_1\):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D_{M_0} - D_{M_1} \sim \chi^2_{df_{M_0} - df_{M_1}} \quad \text{under } H_0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The scaled deviance is \(D^* = D/\phi\), and Pearson's chi-square statistic is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\chi^2_P = \sum_{i=1}^n \frac{(y_i - \hat{\mu}_i)^2}{V(\hat{\mu}_i)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====5) Model Diagnostics====&lt;br /&gt;
Key diagnostic tools:&lt;br /&gt;
* Pearson residuals: \(r_i^P = \frac{y_i - \hat{\mu}_i}{\sqrt{V(\hat{\mu}_i)}}\)&lt;br /&gt;
* Deviance residuals: \(r_i^D = \text{sign}(y_i - \hat{\mu}_i)\sqrt{d_i}\)&lt;br /&gt;
* Leverage: \(h_{ii}\) from the hat matrix \(\mathbf{H} = \mathbf{W}^{1/2}\mathbf{X}(\mathbf{X}^\top\mathbf{W}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{1/2}\)&lt;br /&gt;
* Cook's distance: \(D_i = \frac{r_i^2 h_{ii}}{p(1-h_{ii})}\)&lt;br /&gt;
&lt;br /&gt;
===Applications===&lt;br /&gt;
&lt;br /&gt;
====Example 1: Logistic Regression for Contraceptive Use====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Load and prepare data&lt;br /&gt;
# https://grodri.github.io/datasets/cuse.dat&lt;br /&gt;
cuse &amp;lt;- read.table(&amp;quot;https://grodri.github.io/datasets/cuse.dat&amp;quot;, header=TRUE)&lt;br /&gt;
cat(&amp;quot;First few rows of dataset:\n&amp;quot;)&lt;br /&gt;
print(head(cuse))&lt;br /&gt;
&lt;br /&gt;
# Fit binomial GLM with logit link&lt;br /&gt;
model1 &amp;lt;- glm(cbind(using, notUsing) ~ age + education + wantsMore,&lt;br /&gt;
              family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Confidence Intervals ===\n&amp;quot;)&lt;br /&gt;
confint(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Odds Ratios with 95% CI ===\n&amp;quot;)&lt;br /&gt;
exp_coef &amp;lt;- exp(coef(model1))&lt;br /&gt;
exp_ci &amp;lt;- exp(confint(model1))&lt;br /&gt;
cbind(OR = exp_coef, exp_ci)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Comparison (Likelihood Ratio Test) ===\n&amp;quot;)&lt;br /&gt;
# Reduced model without education&lt;br /&gt;
model_reduced &amp;lt;- glm(cbind(using, notUsing) ~ age + wantsMore,&lt;br /&gt;
                     family = binomial, data = cuse)&lt;br /&gt;
anova(model_reduced, model1, test = &amp;quot;Chisq&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Diagnostic plots&lt;br /&gt;
par(mfrow = c(2, 2))&lt;br /&gt;
plot(model1, which = 1:4)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
'''Result Interpretation''':&lt;br /&gt;
&lt;br /&gt;
Let's explicate ''how to interpret the parameter estimates in this GLM model''.&lt;br /&gt;
Specifically, as this is a bivariate outcome, the estimates are not correlations&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
glm(formula = cbind(using, notUsing) ~ age + education + wantsMore, &lt;br /&gt;
    family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
Coefficients:&lt;br /&gt;
             Estimate Std. Error z value Pr(&amp;gt;|z|)    &lt;br /&gt;
(Intercept)   -0.8082     0.1590  -5.083 3.71e-07 ***&lt;br /&gt;
age25-29       0.3894     0.1759   2.214  0.02681 *  &lt;br /&gt;
age30-39       0.9086     0.1646   5.519 3.40e-08 ***&lt;br /&gt;
age40-49       1.1892     0.2144   5.546 2.92e-08 ***&lt;br /&gt;
educationlow  -0.3250     0.1240  -2.620  0.00879 ** &lt;br /&gt;
wantsMoreyes  -0.8330     0.1175  -7.091 1.33e-12 ***&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
In a ''Generalized Linear Model (GLM)'' with a ''binomial family'' and a ''logit link'' (the default for ''R''), the coefficients represent '''log-odds ratios'''.&lt;br /&gt;
Since our outcome is structured as &lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
   cbind(using, notUsing),&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
we are modeling the probability of &amp;quot;using&amp;quot; (success) versus &amp;quot;not using&amp;quot; (failure).&lt;br /&gt;
&lt;br /&gt;
That is, the model interpretation is in terms of '''Log-Odds'''.&lt;br /&gt;
In this model, the relationship between the ''covariate predictors'' and the ''outcome'' is defined by the ''logit'' function&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\text{logit}(p) = \ln\left(\frac{p}{1-p}\right) = \beta_0 + \beta_1X_1 + \dots + \beta_kX_k.&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The estimate ''Sign'' matters:&lt;br /&gt;
&lt;br /&gt;
: ''Positive Estimate'': As the predictor increases (or if the category is present), the probability of &amp;quot;using&amp;quot; increases.&lt;br /&gt;
: ''Negative Estimate'': As the predictor increases, the probability of &amp;quot;using&amp;quot; decreases.&lt;br /&gt;
&lt;br /&gt;
The estimate ''Magnitude'' is interpreted via the &amp;quot;Exponentiation Trick&amp;quot;.&lt;br /&gt;
Because humans don't think naturally in &amp;quot;log-odds&amp;quot; terms, we usually ''exponentiate'' the coefficients (&amp;lt;math&amp;gt;e^\beta&amp;lt;/math&amp;gt;) to get the ''Odds Ratios (OR)''.&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Variable&lt;br /&gt;
! Estimate (β)&lt;br /&gt;
! Odds Ratio (e&amp;lt;sup&amp;gt;β&amp;lt;/sup&amp;gt;)&lt;br /&gt;
! Interpretation&lt;br /&gt;
|-&lt;br /&gt;
| '''age40-49'''&lt;br /&gt;
| 1.1892&lt;br /&gt;
| ≈ 3.28&lt;br /&gt;
| Women aged 40-49 have '''3.28 times the odds''' of using contraception compared to the reference age group (likely &amp;lt;25).&lt;br /&gt;
|-&lt;br /&gt;
| '''wantsMoreyes'''&lt;br /&gt;
| -0.8330&lt;br /&gt;
| ≈ 0.43&lt;br /&gt;
| Women who want more children have '''57% lower odds''' (1 - 0.43) of using contraception than those who don't.&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
Specific Breakdown of the Results:&lt;br /&gt;
&lt;br /&gt;
* ''Age'' (Categorical): R has treated age as a ''factor''. The baseline (reference) group is the youngest group (''under 25'').&lt;br /&gt;
:: As age increases (&amp;lt;math&amp;gt;25-29 \ \to 30-39\  \to 40-49&amp;lt;/math&amp;gt;), the coefficients become increasingly positive (&amp;lt;math&amp;gt;0.38 \to 0.90 \to 1.18&amp;lt;/math&amp;gt;).&lt;br /&gt;
:: Meaning: Older women in this dataset are significantly more likely to use contraception than the youngest group.&lt;br /&gt;
&lt;br /&gt;
* Education:&lt;br /&gt;
:: Estimate: -0.3250&lt;br /&gt;
:: Meaning: Being in the ''low education'' group is associated with a decrease in the log-odds of using contraception compared to the ''high'' group. Specifically, their odds are about 28\% lower (&amp;lt;math&amp;gt;e^{-0.325} \approx 0.72&amp;lt;/math&amp;gt;).&lt;br /&gt;
&lt;br /&gt;
* Wants More Children:&lt;br /&gt;
:: Estimate: -0.8330&lt;br /&gt;
:: Meaning: This is a strong negative predictor. If a woman wants more children, the probability of her using contraception drops significantly.&lt;br /&gt;
&lt;br /&gt;
We can have a quick &amp;quot;reality check&amp;quot; on the measuring units. Unlike a standard correlation (which is bounded between -1 and 1), these estimates can be any real number.&lt;br /&gt;
&lt;br /&gt;
:: An estimate of 0 means the variable has no effect on the odds (Odds Ratio = 1).&lt;br /&gt;
:: The z-value tells us how many standard errors the estimate is from zero. Since all p-values are small (&amp;lt;math&amp;gt;&amp;lt; 0.05&amp;lt;/math&amp;gt;), all these predictors are statistically significant.&lt;br /&gt;
&lt;br /&gt;
====Example 2: Poisson Regression for Count Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with built-in R dataset: AIDS cases in Belgium&lt;br /&gt;
# Load necessary libraries&lt;br /&gt;
if (!require(&amp;quot;MASS&amp;quot;)) install.packages(&amp;quot;MASS&amp;quot;)&lt;br /&gt;
library(MASS)&lt;br /&gt;
&lt;br /&gt;
# Load AIDS dataset&lt;br /&gt;
data(Aids2)&lt;br /&gt;
cat(&amp;quot;AIDS dataset structure:\n&amp;quot;)&lt;br /&gt;
str(Aids2)&lt;br /&gt;
&lt;br /&gt;
# Prepare data: count of AIDS cases by year and state&lt;br /&gt;
aids_counts &amp;lt;- aggregate(cbind(count = 1:nrow(Aids2)) ~ age + state,&lt;br /&gt;
                         data = Aids2, FUN = length)&lt;br /&gt;
&lt;br /&gt;
# Fit Poisson regression&lt;br /&gt;
model2 &amp;lt;- glm(count ~ age + state, family = poisson, data = aids_counts)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Poisson Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model2)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Check for Overdispersion ===\n&amp;quot;)&lt;br /&gt;
# Pearson chi-square statistic&lt;br /&gt;
pearson_chi2 &amp;lt;- sum(residuals(model2, type = &amp;quot;pearson&amp;quot;)^2)&lt;br /&gt;
df_resid &amp;lt;- df.residual(model2)&lt;br /&gt;
dispersion &amp;lt;- pearson_chi2 / df_resid&lt;br /&gt;
cat(&amp;quot;Pearson χ²:&amp;quot;, pearson_chi2, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;Dispersion parameter:&amp;quot;, dispersion, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;p-value for H0: φ=1:&amp;quot;, pchisq(pearson_chi2, df_resid, lower.tail = FALSE), &amp;quot;\n&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# If overdispersed, fit quasipoisson model&lt;br /&gt;
if (dispersion &amp;gt; 1.5) {&lt;br /&gt;
  cat(&amp;quot;\n=== Fitting Quasi-Poisson Model (accounting for overdispersion) ===\n&amp;quot;)&lt;br /&gt;
  model2_qp &amp;lt;- glm(count ~ age + state, family = quasipoisson, data = aids_counts)&lt;br /&gt;
  summary(model2_qp)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Predict expected counts&lt;br /&gt;
cat(&amp;quot;\n=== Predictions for First 10 Observations ===\n&amp;quot;)&lt;br /&gt;
predictions &amp;lt;- predict(model2, type = &amp;quot;response&amp;quot;, se.fit = TRUE)&lt;br /&gt;
pred_df &amp;lt;- data.frame(&lt;br /&gt;
  Observed = aids_counts$count[1:10],&lt;br /&gt;
  Predicted = predictions$fit[1:10],&lt;br /&gt;
  SE = predictions$se.fit[1:10],&lt;br /&gt;
  Lower = predictions$fit[1:10] - 1.96 * predictions$se.fit[1:10],&lt;br /&gt;
  Upper = predictions$fit[1:10] + 1.96 * predictions$se.fit[1:10]&lt;br /&gt;
)&lt;br /&gt;
print(pred_df)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Example 3: Gamma Regression for Positive Continuous Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with insurance claims data&lt;br /&gt;
if (!require(&amp;quot;insuranceData&amp;quot;)) install.packages(&amp;quot;insuranceData&amp;quot;)&lt;br /&gt;
library(insuranceData)&lt;br /&gt;
data(dataCar)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;Car insurance claims dataset:\n&amp;quot;)&lt;br /&gt;
str(dataCar)&lt;br /&gt;
&lt;br /&gt;
# Filter for positive claim amounts&lt;br /&gt;
claims_positive &amp;lt;- subset(dataCar, claimcst0 &amp;gt; 0)&lt;br /&gt;
&lt;br /&gt;
# Fit Gamma GLM with log link (common for monetary amounts)&lt;br /&gt;
model3 &amp;lt;- glm(claimcst0 ~ agecat + area + veh_age,&lt;br /&gt;
              family = Gamma(link = &amp;quot;log&amp;quot;),&lt;br /&gt;
              data = claims_positive)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Gamma Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model3)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Checking Gamma Model Assumptions ===\n&amp;quot;)&lt;br /&gt;
# Check residuals&lt;br /&gt;
res_gamma &amp;lt;- residuals(model3, type = &amp;quot;deviance&amp;quot;)&lt;br /&gt;
par(mfrow = c(1, 2))&lt;br /&gt;
hist(res_gamma, main = &amp;quot;Deviance Residuals&amp;quot;, xlab = &amp;quot;Residuals&amp;quot;)&lt;br /&gt;
qqnorm(res_gamma, main = &amp;quot;Q-Q Plot of Residuals&amp;quot;)&lt;br /&gt;
qqline(res_gamma)&lt;br /&gt;
&lt;br /&gt;
# Scale-location plot&lt;br /&gt;
fitted_values &amp;lt;- fitted(model3)&lt;br /&gt;
plot(fitted_values, sqrt(abs(res_gamma)),&lt;br /&gt;
     xlab = &amp;quot;Fitted Values&amp;quot;, ylab = &amp;quot;√|Deviance Residuals|&amp;quot;,&lt;br /&gt;
     main = &amp;quot;Scale-Location Plot&amp;quot;)&lt;br /&gt;
lines(lowess(fitted_values, sqrt(abs(res_gamma))), col = &amp;quot;red&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Software Implementation in R===&lt;br /&gt;
&lt;br /&gt;
Note that `family` is a function (a &amp;quot;closure&amp;quot;), not a list or an object with a \(\$family\) component. In thecode, we're passing&lt;br /&gt;
the family function itself (e.g., gaussian) to the \(run\_glm\_analysis()\) function, and later accessing \(family\$family\), &lt;br /&gt;
which doesn’t exist—because gaussian is a function, not a fitted model object.&lt;br /&gt;
In R, gaussian, binomial, poisson, etc., are functions that return family objects when called.&lt;br /&gt;
We're passing the function, not the result of calling it.&lt;br /&gt;
However, `glm()` internally calls `family()`, i.e., `gaussian()`, to get the actual family list object, which does have a \(\$family\) element.&lt;br /&gt;
Instead of checking \(family\$family\), we extract the family name from the ''fitted model'', not from the input argument.&lt;br /&gt;
Specifically, after fitting the model, we use \(model\$family\$family\), which is a character string like &amp;quot;gaussian&amp;quot;, &amp;quot;binomial&amp;quot;, etc.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Comprehensive GLM analysis function&lt;br /&gt;
run_glm_analysis &amp;lt;- function(formula, data, family, link = NULL) {&lt;br /&gt;
  # Fit the model&lt;br /&gt;
  if (!is.null(link)) {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family(link = link))&lt;br /&gt;
  } else {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family)&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  # Model summary&lt;br /&gt;
  cat(&amp;quot;=== MODEL SUMMARY ===\n&amp;quot;)&lt;br /&gt;
  print(summary(model))&lt;br /&gt;
  &lt;br /&gt;
  # Confidence intervals&lt;br /&gt;
  cat(&amp;quot;\n=== 95% CONFIDENCE INTERVALS ===\n&amp;quot;)&lt;br /&gt;
  print(confint(model))&lt;br /&gt;
  &lt;br /&gt;
  # Goodness-of-fit tests&lt;br /&gt;
  cat(&amp;quot;\n=== GOODNESS-OF-FIT ===\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Null Deviance:&amp;quot;, model$null.deviance, &amp;quot;on&amp;quot;, model$df.null, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Residual Deviance:&amp;quot;, model$deviance, &amp;quot;on&amp;quot;, model$df.residual, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;AIC:&amp;quot;, AIC(model), &amp;quot;\n&amp;quot;)&lt;br /&gt;
  &lt;br /&gt;
  # Check for overdispersion (for Poisson and binomial families)&lt;br /&gt;
  family_name &amp;lt;- model$family$family  # ✅ Extract from fitted model&lt;br /&gt;
  if (family_name %in% c(&amp;quot;poisson&amp;quot;, &amp;quot;binomial&amp;quot;, &amp;quot;quasipoisson&amp;quot;, &amp;quot;quasibinomial&amp;quot;)) {&lt;br /&gt;
    cat(&amp;quot;\n=== OVERDISPERSION CHECK ===\n&amp;quot;)&lt;br /&gt;
    dispersion &amp;lt;- model$deviance / model$df.residual&lt;br /&gt;
    cat(&amp;quot;Dispersion parameter:&amp;quot;, round(dispersion, 4), &amp;quot;\n&amp;quot;)&lt;br /&gt;
    if (abs(dispersion - 1) &amp;gt; 0.1) {&lt;br /&gt;
      direction &amp;lt;- ifelse(dispersion &amp;gt; 1, &amp;quot;over&amp;quot;, &amp;quot;under&amp;quot;)&lt;br /&gt;
      cat(&amp;quot;Note: Significant&amp;quot;, direction, &amp;quot;dispersion detected\n&amp;quot;)&lt;br /&gt;
    }&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  return(model)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Example usage with mtcars dataset&lt;br /&gt;
cat(&amp;quot;\n\n=== EXAMPLE: GAUSSIAN GLM (equivalent to linear regression) ===\n&amp;quot;)&lt;br /&gt;
data(mtcars)&lt;br /&gt;
model_gaussian &amp;lt;- run_glm_analysis(&lt;br /&gt;
  formula = mpg ~ wt + hp + cyl,&lt;br /&gt;
  data = mtcars,&lt;br /&gt;
  family = gaussian,&lt;br /&gt;
  link = &amp;quot;identity&amp;quot;&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Compare with lm() for verification&lt;br /&gt;
cat(&amp;quot;\n=== COMPARISON WITH lm() ===\n&amp;quot;)&lt;br /&gt;
model_lm &amp;lt;- lm(mpg ~ wt + hp + cyl, data = mtcars)&lt;br /&gt;
print(summary(model_lm))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Common GLM Families and Links in R===&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Family !! Default Link !! Alternative Links !! Typical Use&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || Continuous, symmetric data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;binomial()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;logit&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;probit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cauchit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cloglog&amp;lt;/code&amp;gt; || Binary/count proportions&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;poisson()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;sqrt&amp;lt;/code&amp;gt; || Count data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;Gamma()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, right-skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;inverse.gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;1/μ²&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, highly skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;quasi()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || User-defined || Overdispersed data&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Practical Considerations===&lt;br /&gt;
&lt;br /&gt;
* Model Selection:&lt;br /&gt;
** AIC (Akaike Information Criterion): \( \text{AIC} = -2\ell + 2p \)&lt;br /&gt;
** BIC (Bayesian Information Criterion): \( \text{BIC} = -2\ell + p\log(n) \)&lt;br /&gt;
** Cross-validation: Particularly useful for predictive performance.&lt;br /&gt;
&lt;br /&gt;
* Handling Overdispersion: For Poisson models: \( \text{Var}(Y) = \phi\mu \) where \(\phi &amp;gt; 1\) indicates overdispersion&lt;br /&gt;
Solutions:&lt;br /&gt;
** Use quasi-Poisson model&lt;br /&gt;
** Use negative binomial distribution&lt;br /&gt;
** Include random effects.&lt;br /&gt;
&lt;br /&gt;
* Zero-Inflation: For count data with excess zeros, consider:&lt;br /&gt;
** Zero-inflated Poisson (ZIP) model&lt;br /&gt;
** Zero-inflated negative binomial (ZINB) model&lt;br /&gt;
** Hurdle models.&lt;br /&gt;
&lt;br /&gt;
===Advanced Topics===&lt;br /&gt;
&lt;br /&gt;
==== Mixed Effects GLM (GLMM)====&lt;br /&gt;
&lt;br /&gt;
Extension incorporating random effects:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
g(E[Y|b]) = \mathbf{X}\beta + \mathbf{Z}b&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \( b \sim N(0, \mathbf{G}) \).&lt;br /&gt;
&lt;br /&gt;
Implementation in R with &amp;lt;code&amp;gt;lme4::glmer()&amp;lt;/code&amp;gt;:&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(lme4)&lt;br /&gt;
&lt;br /&gt;
# generate random data&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
id &amp;lt;- seq(n)&lt;br /&gt;
day &amp;lt;- 1:20&lt;br /&gt;
mydata &amp;lt;- expand.grid(id = id, day = day)&lt;br /&gt;
set.seed(1)&lt;br /&gt;
trt &amp;lt;- sample(c(&amp;quot;control&amp;quot;, &amp;quot;treat&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
sex &amp;lt;- sample(c(&amp;quot;female&amp;quot;, &amp;quot;male&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
mydata$trt &amp;lt;- trt[mydata$id]&lt;br /&gt;
mydata$sex &amp;lt;- sex[mydata$id]&lt;br /&gt;
mydata &amp;lt;- mydata[order(mydata$id, mydata$day),]&lt;br /&gt;
rownames(mydata) &amp;lt;- NULL&lt;br /&gt;
head(mydata, n = 10)&lt;br /&gt;
&lt;br /&gt;
mydata$trtsex &amp;lt;- interaction(mydata$trt, mydata$sex)&lt;br /&gt;
probs &amp;lt;- c(0.40, 0.85, 0.30, 0.50)&lt;br /&gt;
names(probs) &amp;lt;- levels(mydata$trtsex)&lt;br /&gt;
mydata$p &amp;lt;- probs[mydata$trtsex]&lt;br /&gt;
&lt;br /&gt;
set.seed(3)&lt;br /&gt;
r_probs &amp;lt;- rnorm(n = n, mean = 0, sd = 0.03)&lt;br /&gt;
mydata$random_p &amp;lt;- r_probs[mydata$id]&lt;br /&gt;
mydata$p &amp;lt;- mydata$p + mydata$random_p&lt;br /&gt;
&lt;br /&gt;
# use the probabilities to generate zeroes and ones with the binom() function.&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n = nrow(mydata), size = 1, prob = mydata$p)&lt;br /&gt;
&lt;br /&gt;
# using the sim data, inspect the first few records.&lt;br /&gt;
&lt;br /&gt;
head(mydata[c(&amp;quot;id&amp;quot;, &amp;quot;day&amp;quot;, &amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;, &amp;quot;p&amp;quot;, &amp;quot;y&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
# Example with binary outcome and random intercept&lt;br /&gt;
m &amp;lt;- glmer(y ~ trt * sex + (1|id), data = mydata, family = binomial)&lt;br /&gt;
summary(m, corr = FALSE)&lt;br /&gt;
&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Bayesian GLM====&lt;br /&gt;
&lt;br /&gt;
Using Markov Chain Monte Carlo (MCMC) methods.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
# Bayesian logistic regression&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(y ~ day +  trt + sex + p, # predictors,&lt;br /&gt;
                        family = binomial,&lt;br /&gt;
                        data = mydata,&lt;br /&gt;
                        prior = normal(0, 2.5),&lt;br /&gt;
                        prior_intercept = normal(0, 5))&lt;br /&gt;
&lt;br /&gt;
print(bayes_model, digits = 2)&lt;br /&gt;
&lt;br /&gt;
# Coefficient Plot (Forest Plot)&lt;br /&gt;
&lt;br /&gt;
plot(bayes_model, &amp;quot;areas&amp;quot;)  # density + intervals&lt;br /&gt;
# OR more customizable:&lt;br /&gt;
library(bayesplot)&lt;br /&gt;
mcmc_intervals(bayes_model, prob = 0.95) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Distributions of Coefficients&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Posterior Predictive Checks (PPC)&lt;br /&gt;
# Check if the model can reproduce data like the observed&lt;br /&gt;
pp_check(bayes_model, plotfun = &amp;quot;hist&amp;quot;, nreps = 100) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Predictive Check (Histogram of y_rep vs y)&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Another Bayesian Experiment.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
set.seed(123)&lt;br /&gt;
&lt;br /&gt;
# Simulate data — explicitly make trt and sex into factors&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
mydata &amp;lt;- data.frame(&lt;br /&gt;
  day = sample(1:30, n, replace = TRUE),&lt;br /&gt;
  trt = factor(sample(c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;)),&lt;br /&gt;
  sex = factor(sample(c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;)),&lt;br /&gt;
  p   = rnorm(n, mean = 50, sd = 10)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# True log-odds&lt;br /&gt;
logit_p &amp;lt;- -1 - 0.02 * mydata$day + &lt;br /&gt;
           0.6 * (as.numeric(mydata$trt) - 1) +   # A=0, B=1&lt;br /&gt;
           0.3 * (as.numeric(mydata$sex) - 1) +    # F=0, M=1&lt;br /&gt;
           0.04 * mydata$p&lt;br /&gt;
&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n, size = 1, prob = plogis(logit_p))&lt;br /&gt;
&lt;br /&gt;
# Fit model&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(&lt;br /&gt;
  y ~ day + trt + sex + p,&lt;br /&gt;
  family = binomial,&lt;br /&gt;
  data = mydata,&lt;br /&gt;
  prior = normal(0, 2.5),&lt;br /&gt;
  prior_intercept = normal(0, 5),&lt;br /&gt;
  seed = 42,&lt;br /&gt;
  refresh = 0&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Create New Data&lt;br /&gt;
# Confirm variables are factors&lt;br /&gt;
str(mydata[c(&amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
# Create prediction grid&lt;br /&gt;
newdata &amp;lt;- expand.grid(&lt;br /&gt;
  day = seq(min(mydata$day), max(mydata$day), length.out = 30),&lt;br /&gt;
  trt = levels(mydata$trt)[1],      # e.g., &amp;quot;A&amp;quot;&lt;br /&gt;
  sex = levels(mydata$sex)[1],      # e.g., &amp;quot;F&amp;quot;&lt;br /&gt;
  p   = median(mydata$p)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Verify it worked&lt;br /&gt;
stopifnot(nrow(newdata) &amp;gt; 0)&lt;br /&gt;
str(newdata)&lt;br /&gt;
&lt;br /&gt;
# Generate Predictions from the Bayesian Posterior Probability (after fitting the Bayesian Model):&lt;br /&gt;
# Posterior predicted probabilities&lt;br /&gt;
post_pred &amp;lt;- posterior_linpred(bayes_model, newdata = newdata, transform = TRUE)&lt;br /&gt;
&lt;br /&gt;
# Summarize&lt;br /&gt;
pred_summary &amp;lt;- data.frame(&lt;br /&gt;
  day = newdata$day,&lt;br /&gt;
  prob = apply(post_pred, 2, median),&lt;br /&gt;
  lower = apply(post_pred, 2, quantile, 0.025),&lt;br /&gt;
  upper = apply(post_pred, 2, quantile, 0.975)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Plot&lt;br /&gt;
library(ggplot2)&lt;br /&gt;
ggplot(pred_summary, aes(x = day, y = prob)) +&lt;br /&gt;
  geom_line(color = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  geom_ribbon(aes(ymin = lower, ymax = upper), alpha = 0.2, fill = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  labs(&lt;br /&gt;
    title = &amp;quot;Predicted Probability of Outcome by Day&amp;quot;,&lt;br /&gt;
    subtitle = &amp;quot;Holding treatment = A, sex = F, and p = median&amp;quot;,&lt;br /&gt;
    y = &amp;quot;P(y = 1)&amp;quot;,&lt;br /&gt;
    x = &amp;quot;Day&amp;quot;&lt;br /&gt;
  ) +&lt;br /&gt;
  theme_minimal()&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==== ''Gamma GLM'' vs. ''Beta Regression''====&lt;br /&gt;
&lt;br /&gt;
The ''Gamma GLM'' and ''Beta Regression'' models are designed for very different types of &lt;br /&gt;
&amp;quot;continuous&amp;quot; data.&lt;br /&gt;
The fundamental difference lies in the '''domain''' (the range of possible values) and the &lt;br /&gt;
'''variance structure''' of the data they expect.&lt;br /&gt;
&lt;br /&gt;
===== The Gamma Model (''glm'')=====&lt;br /&gt;
&lt;br /&gt;
The Gamma distribution is used for ''strictly positive, continuous data'' &amp;lt;math&amp;gt;(0, \infty)&amp;lt;/math&amp;gt;. It is appropriate for data that is &amp;quot;skewed&amp;quot; to the right, where the variance increases as the mean increases.&lt;br /&gt;
If a variable, like ''qol_score'' is a raw score (e.g., 0 to 100), the link, ''link = &amp;quot;log&amp;quot;'',&lt;br /&gt;
ensures the predicted values remain positive.&lt;br /&gt;
&lt;br /&gt;
In a Gamma GLM with a log link, we model the mean &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt; as&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\ln(\mu_i) = \beta_0 + \beta_1 X_1 + \beta_2 X_2 + \dots + \beta_k X_k.&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The probability density function (PDF) for the Gamma distribution is&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;f(y; \alpha, \beta) = \frac{\beta^\alpha y^{\alpha-1} e^{-\beta y}}{\Gamma(\alpha)} \quad \text{for } y &amp;gt; 0.&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===== The Beta Model (''betareg'')=====&lt;br /&gt;
&lt;br /&gt;
The Beta distribution, is used specifically for '''rates, proportions, or scores constrained between 0 and 1''', &amp;lt;math&amp;gt;(0, 1)&amp;lt;/math&amp;gt;. Unlike the Gamma model, which can go off to infinity, the Beta distribution is &amp;quot;boxed in&amp;quot; this interval.&lt;br /&gt;
For instance, if ''qol_scaled'' is normalized, e.g., by dividing QoL score by its maximum value,&lt;br /&gt;
it's values are forced into &amp;lt;math&amp;gt;(0, 1)&amp;lt;/math&amp;gt; interval.&lt;br /&gt;
Beta distribution shape is veryflexible. It can be U-shaped, J-shaped, or symmetric, making it perfect for &amp;quot;ceiling&amp;quot; or &amp;quot;floor&amp;quot; effects in Quality of Life scores.&lt;br /&gt;
&lt;br /&gt;
Beta regression typically uses a '''logit link''' by default to keep predictions between 0 and 1:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\ln\left(\frac{\mu_i}{1-\mu_i}\right) = \beta_0 + \beta_1 X_1 + \dots + \beta_k X_k.&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The PDF for the Beta distribution (parameterized by mean &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt; and precision &amp;lt;math&amp;gt;\phi&amp;lt;/math&amp;gt;) is:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;f(y; \mu, \phi) = \frac{\Gamma(\phi)}{\Gamma(\mu\phi)\Gamma((1-\mu)\phi)} y^{\mu\phi-1} (1-y)^{(1-\mu)\phi-1} \quad \text{for } y \in (0, 1).&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Key Differences between ''Gamma GLM'' and ''Beta Regression''&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Feature&lt;br /&gt;
! Gamma Model (model4)&lt;br /&gt;
! Beta Model (model_beta)&lt;br /&gt;
|-&lt;br /&gt;
| '''Outcome Range'''&lt;br /&gt;
| &amp;lt;math&amp;gt;(0, \infty)&amp;lt;/math&amp;gt; (Positive numbers)&lt;br /&gt;
| &amp;lt;math&amp;gt;(0, 1)&amp;lt;/math&amp;gt; (Proportions/Scaled)&lt;br /&gt;
|-&lt;br /&gt;
| '''Typical Use'''&lt;br /&gt;
| Costs, rainfall, raw test scores&lt;br /&gt;
| Percentages, scaled QoL indices&lt;br /&gt;
|-&lt;br /&gt;
| '''Interpretation'''&lt;br /&gt;
| Log-link: Coefficients are multiplicative&lt;br /&gt;
| Logit-link: Coefficients are odds ratios&lt;br /&gt;
|-&lt;br /&gt;
| '''Boundaries'''&lt;br /&gt;
| Can handle very large values&lt;br /&gt;
| Naturally handles &amp;quot;ceiling effects&amp;quot;&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
If the outcome dependent variable (DV) qol_score is a raw sum that could theoretically be much higher, stick with Gamma. But when ''qol_score'' is a bounded instrument (like a visual analog scale from 0–1) with many observations scoring near the very top or bottom, ''Beta'' may be superior because it understands the &amp;quot;walls&amp;quot; of the scale.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(readr)&lt;br /&gt;
 &lt;br /&gt;
 # REMOTE URL&lt;br /&gt;
 url_wide &amp;lt;- &amp;quot;https://socr.umich.edu/docs/uploads/2026/SOCR_CRDS_CompAnalysis_CaseStudy_wide.csv&amp;quot;&lt;br /&gt;
 url_long &amp;lt;- &amp;quot;https://socr.umich.edu/docs/uploads/2026/SOCR_CRDS_CompAnalysis_CaseStudy_long.csv&amp;quot;&lt;br /&gt;
 &lt;br /&gt;
 # Import: Reading the Wide data&lt;br /&gt;
 df_wide &amp;lt;- read_csv(url_wide, col_types = cols(&lt;br /&gt;
     patient_id      = col_factor(),    # Prevents math on IDs&lt;br /&gt;
     age             = col_double(),&lt;br /&gt;
     gender          = col_character(),&lt;br /&gt;
     treatment       = col_character(),&lt;br /&gt;
     smoking_history = col_character(),&lt;br /&gt;
     baseline_fvc    = col_double(),&lt;br /&gt;
     fvc_0           = col_double(),&lt;br /&gt;
     fvc_6           = col_double(),&lt;br /&gt;
     fvc_12          = col_double(),&lt;br /&gt;
     qol_score       = col_double(),&lt;br /&gt;
     walk_dist       = col_double(),&lt;br /&gt;
     success         = col_integer(),   # 0 or 1 is best as integer&lt;br /&gt;
     hospital_visits = col_integer(),&lt;br /&gt;
     physician_notes = col_character()&lt;br /&gt;
 ))&lt;br /&gt;
                                                                                   &lt;br /&gt;
 # Import: Reading the Long data&lt;br /&gt;
 df_long &amp;lt;- read_csv(url_long, col_types = cols(&lt;br /&gt;
     patient_id = col_factor(),  &lt;br /&gt;
     age = col_double(),&lt;br /&gt;
     gender = col_character(),&lt;br /&gt;
     timepoint = col_factor(levels = c(&amp;quot;Month 0&amp;quot;, &amp;quot;Month 6&amp;quot;, &amp;quot;Month 12&amp;quot;))&lt;br /&gt;
 ))&lt;br /&gt;
&lt;br /&gt;
df_wide$qol_scaled &amp;lt;- df_wide$qol_score / (max(df_wide$qol_score) + 0.0001) &lt;br /&gt;
 &lt;br /&gt;
 # Use the betareg package&lt;br /&gt;
 library(betareg)&lt;br /&gt;
&lt;br /&gt;
model4 &amp;lt;- glm(qol_score ~ treatment + gender + age + baseline_fvc + success, family = Gamma(link = &amp;quot;log&amp;quot;), data = df_wide)&lt;br /&gt;
summary(model4)&lt;br /&gt;
model_beta &amp;lt;- betareg(qol_scaled ~ treatment + gender + baseline_fvc + success + age, data = df_wide)&lt;br /&gt;
summary(model_beta )&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Problems===&lt;br /&gt;
&lt;br /&gt;
1. Conceptual Exercises:&lt;br /&gt;
&lt;br /&gt;
   a) Derive the score equations for a Poisson GLM with log link&lt;br /&gt;
   b) Show that the binomial distribution with logit link is the canonical link&lt;br /&gt;
   c) Prove that the deviance for a normal GLM equals the residual sum of squares&lt;br /&gt;
&lt;br /&gt;
2. Applied Problems:&lt;br /&gt;
&lt;br /&gt;
   a) Analyze the [http://wiki.socr.umich.edu/index.php/SOCR_Data_Dinov_021808_ConsumerPriceIndex3Way Consumer Price Index] data using appropriate GLM&lt;br /&gt;
   b) Model the [http://wiki.socr.umich.edu/index.php/SOCR_Data_MonetaryBase1959_2009 Monetary Base] data considering temporal autocorrelation&lt;br /&gt;
   c) Using the &amp;lt;code&amp;gt;iris&amp;lt;/code&amp;gt; dataset, build a multinomial logistic regression to classify species&lt;br /&gt;
&lt;br /&gt;
3. Practice Simulation Study:&lt;br /&gt;
&lt;br /&gt;
   &amp;lt;pre&amp;gt;&lt;br /&gt;
   # Simulate data from a logistic regression model&lt;br /&gt;
   set.seed(123)&lt;br /&gt;
   n &amp;lt;- 1000&lt;br /&gt;
   x1 &amp;lt;- rnorm(n)&lt;br /&gt;
   x2 &amp;lt;- rnorm(n)&lt;br /&gt;
   beta &amp;lt;- c(0.5, 1, -0.5)&lt;br /&gt;
   linear_predictor &amp;lt;- beta[1] + beta[2]*x1 + beta[3]*x2&lt;br /&gt;
   probabilities &amp;lt;- plogis(linear_predictor)&lt;br /&gt;
   y &amp;lt;- rbinom(n, size = 1, prob = probabilities)&lt;br /&gt;
   &lt;br /&gt;
   # Fit model and evaluate performance&lt;br /&gt;
   sim_data &amp;lt;- data.frame(y = y, x1 = x1, x2 = x2)&lt;br /&gt;
   model_sim &amp;lt;- glm(y ~ x1 + x2, family = binomial, data = sim_data)&lt;br /&gt;
   &lt;br /&gt;
   # Calculate bias and MSE&lt;br /&gt;
   beta_hat &amp;lt;- coef(model_sim)&lt;br /&gt;
   bias &amp;lt;- beta_hat - beta&lt;br /&gt;
   mse &amp;lt;- mean((beta_hat - beta)^2)&lt;br /&gt;
   &lt;br /&gt;
   cat(&amp;quot;True parameters:&amp;quot;, beta, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Estimated parameters:&amp;quot;, beta_hat, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Bias:&amp;quot;, bias, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;MSE:&amp;quot;, mse, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   &amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
&lt;br /&gt;
1. McCullagh, P., &amp;amp; Nelder, J. A. (1989). *Generalized Linear Models* (2nd ed.). Chapman and Hall.&lt;br /&gt;
&lt;br /&gt;
2. Dobson, A. J., &amp;amp; Barnett, A. G. (2018). *An Introduction to Generalized Linear Models* (4th ed.). CRC Press.&lt;br /&gt;
&lt;br /&gt;
3. Agresti, A. (2015). *Foundations of Linear and Generalized Linear Models*. Wiley.&lt;br /&gt;
&lt;br /&gt;
4. Fahrmeir, L., Kneib, T., Lang, S., &amp;amp; Marx, B. (2013). *Regression: Models, Methods and Applications*. Springer.&lt;br /&gt;
&lt;br /&gt;
5. Wood, S. N. (2017). *Generalized Additive Models: An Introduction with R* (2nd ed.). Chapman and Hall/CRC.&lt;br /&gt;
&lt;br /&gt;
====Online Resources====&lt;br /&gt;
* [https://en.wikipedia.org/wiki/Generalized_linear_model GLM Wikipedia]&lt;br /&gt;
* [https://data.princeton.edu/wws509/notes/a2.pdf GLM Theory Notes]&lt;br /&gt;
* [https://www.jstatsoft.org/article/view/v015i12 GLM in R Tutorial]&lt;br /&gt;
* [https://cran.r-project.org/web/views/SocialSciences.html R Resources for Social Sciences]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php?title=SMHS_GLM}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18354</id>
		<title>SMHS GLM</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18354"/>
		<updated>2026-03-13T21:08:19Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Advanced Topics */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Generalized Linear Modeling (GLM) ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Generalized Linear Modeling (GLM) is a flexible generalization of ordinary linear regression that allows response variables to have error distribution models other than a normal distribution. GLM extends linear regression by allowing the linear model to be related to the response variable via a link function and enabling the variance of each measurement to be a function of its predicted value. This framework unifies statistical models including linear regression, logistic regression, and Poisson regression. Estimation methods include iteratively reweighted least squares for maximum likelihood estimation, Bayesian approaches, and least squares fitted to variance-stabilized responses.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
While linear regression models linear relationships between response and predictors, many real-world scenarios involve response variables that don't follow normal distributions. For example:&lt;br /&gt;
* Binary outcomes (yes/no decisions) with probabilities bounded between 0 and 1&lt;br /&gt;
* Count data (number of events) that follow Poisson distributions&lt;br /&gt;
* Survival times that follow exponential or Weibull distributions&lt;br /&gt;
&lt;br /&gt;
GLM provides a unified framework for these situations by allowing response variables from the exponential family of distributions.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====1) GLM Components====&lt;br /&gt;
A GLM consists of three components:&lt;br /&gt;
&lt;br /&gt;
1. Random Component: The response variable \(Y\) follows a distribution from the exponential family:&lt;br /&gt;
   &amp;lt;math&amp;gt;f_Y(y|\theta,\phi) = \exp\left\{\frac{y\theta - b(\theta)}{a(\phi)} + c(y,\phi)\right\}&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;\theta&amp;lt;/math&amp;gt; is the natural parameter and \(\phi\) is the dispersion parameter.&lt;br /&gt;
&lt;br /&gt;
2. Systematic Component: The linear predictor \(\eta\):&lt;br /&gt;
   &amp;lt;math&amp;gt;\eta = \mathbf{X}\beta = \beta_0 + \beta_1X_1 + \beta_2X_2 + \cdots + \beta_pX_p&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
3. Link Function: \(g(\cdot)\) that relates the mean \(\mu = E[Y]\) to the linear predictor:&lt;br /&gt;
   &amp;lt;math&amp;gt;g(\mu) = \eta \quad \text{or equivalently} \quad \mu = g^{-1}(\eta)&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The variance function relates the variance to the mean: &amp;lt;math&amp;gt;\text{Var}(Y) = a(\phi)V(\mu)&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;V(\mu)&amp;lt;/math&amp;gt; is the variance function specific to the distribution.&lt;br /&gt;
&lt;br /&gt;
====2) Exponential Family Distributions====&lt;br /&gt;
The exponential family includes many common distributions:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Distribution !! Support !! Natural Parameter \(\theta\) !! \(b(\theta)\) !! Canonical Link \(g(\mu)\) !! Variance Function \(V(\mu)\)&lt;br /&gt;
|-&lt;br /&gt;
| Normal || \(\mathbb{R}\) || \(\mu\) || \(\frac{\theta^2}{2}\) || Identity: \(\mu\) || 1&lt;br /&gt;
|-&lt;br /&gt;
| Binomial || \(\{0,1,\ldots,n\}\) || \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(n\log(1+e^\theta)\) || Logit: \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(\mu\left(1-\frac{\mu}{n}\right)\)&lt;br /&gt;
|-&lt;br /&gt;
| Poisson || \(\mathbb{N}_0\) || \(\log(\mu)\) || \(e^\theta\) || Log: \(\log(\mu)\) || \(\mu\)&lt;br /&gt;
|-&lt;br /&gt;
| Gamma || \(\mathbb{R}^+\) || \(-\frac{1}{\mu}\) || \(-\log(-\theta)\) || Inverse: \(\frac{1}{\mu}\) || \(\mu^2\)&lt;br /&gt;
|-&lt;br /&gt;
| Inverse Gaussian || \(\mathbb{R}^+\) || \(-\frac{1}{2\mu^2}\) || \(-\sqrt{-2\theta}\) || Inverse squared: \(\frac{1}{\mu^2}\) || \(\mu^3\)&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====3) Maximum Likelihood Estimation====&lt;br /&gt;
For a GLM with \(n\) independent observations, the log-likelihood is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\ell(\beta) = \sum_{i=1}^n \frac{y_i\theta_i - b(\theta_i)}{a(\phi)} + c(y_i,\phi)&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\theta_i = \theta(\mu_i)\) and \(\mu_i = g^{-1}(\mathbf{x}_i^\top\beta)\).&lt;br /&gt;
&lt;br /&gt;
The score equations are:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathbf{U}(\beta) = \frac{\partial\ell}{\partial\beta} = \mathbf{X}^\top\mathbf{W}(\mathbf{y} - \boldsymbol{\mu}) = 0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\mathbf{W} = \text{diag}\left\{\frac{1}{a(\phi)V(\mu_i)[g'(\mu_i)]^2}\right\}\).&lt;br /&gt;
&lt;br /&gt;
The Fisher information matrix is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathcal{I}(\beta) = E\left[-\frac{\partial^2\ell}{\partial\beta\partial\beta^\top}\right] = \mathbf{X}^\top\mathbf{W}\mathbf{X}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Parameters are estimated via Iteratively Reweighted Least Squares (IRLS):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\beta^{(t+1)} = \beta^{(t)} + (\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{z}^{(t)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(z_i^{(t)} = \eta_i^{(t)} + (y_i - \mu_i^{(t)})g'(\mu_i^{(t)})\).&lt;br /&gt;
&lt;br /&gt;
====4) Deviance and Goodness-of-Fit====&lt;br /&gt;
The deviance measures goodness-of-fit:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D = 2[\ell(\text{saturated model}) - \ell(\text{fitted model})]&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
For nested models \(M_0 \subset M_1\):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D_{M_0} - D_{M_1} \sim \chi^2_{df_{M_0} - df_{M_1}} \quad \text{under } H_0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The scaled deviance is \(D^* = D/\phi\), and Pearson's chi-square statistic is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\chi^2_P = \sum_{i=1}^n \frac{(y_i - \hat{\mu}_i)^2}{V(\hat{\mu}_i)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====5) Model Diagnostics====&lt;br /&gt;
Key diagnostic tools:&lt;br /&gt;
* Pearson residuals: \(r_i^P = \frac{y_i - \hat{\mu}_i}{\sqrt{V(\hat{\mu}_i)}}\)&lt;br /&gt;
* Deviance residuals: \(r_i^D = \text{sign}(y_i - \hat{\mu}_i)\sqrt{d_i}\)&lt;br /&gt;
* Leverage: \(h_{ii}\) from the hat matrix \(\mathbf{H} = \mathbf{W}^{1/2}\mathbf{X}(\mathbf{X}^\top\mathbf{W}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{1/2}\)&lt;br /&gt;
* Cook's distance: \(D_i = \frac{r_i^2 h_{ii}}{p(1-h_{ii})}\)&lt;br /&gt;
&lt;br /&gt;
===Applications===&lt;br /&gt;
&lt;br /&gt;
====Example 1: Logistic Regression for Contraceptive Use====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Load and prepare data&lt;br /&gt;
# https://grodri.github.io/datasets/cuse.dat&lt;br /&gt;
cuse &amp;lt;- read.table(&amp;quot;https://grodri.github.io/datasets/cuse.dat&amp;quot;, header=TRUE)&lt;br /&gt;
cat(&amp;quot;First few rows of dataset:\n&amp;quot;)&lt;br /&gt;
print(head(cuse))&lt;br /&gt;
&lt;br /&gt;
# Fit binomial GLM with logit link&lt;br /&gt;
model1 &amp;lt;- glm(cbind(using, notUsing) ~ age + education + wantsMore,&lt;br /&gt;
              family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Confidence Intervals ===\n&amp;quot;)&lt;br /&gt;
confint(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Odds Ratios with 95% CI ===\n&amp;quot;)&lt;br /&gt;
exp_coef &amp;lt;- exp(coef(model1))&lt;br /&gt;
exp_ci &amp;lt;- exp(confint(model1))&lt;br /&gt;
cbind(OR = exp_coef, exp_ci)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Comparison (Likelihood Ratio Test) ===\n&amp;quot;)&lt;br /&gt;
# Reduced model without education&lt;br /&gt;
model_reduced &amp;lt;- glm(cbind(using, notUsing) ~ age + wantsMore,&lt;br /&gt;
                     family = binomial, data = cuse)&lt;br /&gt;
anova(model_reduced, model1, test = &amp;quot;Chisq&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Diagnostic plots&lt;br /&gt;
par(mfrow = c(2, 2))&lt;br /&gt;
plot(model1, which = 1:4)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
'''Result Interpretation''':&lt;br /&gt;
&lt;br /&gt;
Let's explicate ''how to interpret the parameter estimates in this GLM model''.&lt;br /&gt;
Specifically, as this is a bivariate outcome, the estimates are not correlations&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
glm(formula = cbind(using, notUsing) ~ age + education + wantsMore, &lt;br /&gt;
    family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
Coefficients:&lt;br /&gt;
             Estimate Std. Error z value Pr(&amp;gt;|z|)    &lt;br /&gt;
(Intercept)   -0.8082     0.1590  -5.083 3.71e-07 ***&lt;br /&gt;
age25-29       0.3894     0.1759   2.214  0.02681 *  &lt;br /&gt;
age30-39       0.9086     0.1646   5.519 3.40e-08 ***&lt;br /&gt;
age40-49       1.1892     0.2144   5.546 2.92e-08 ***&lt;br /&gt;
educationlow  -0.3250     0.1240  -2.620  0.00879 ** &lt;br /&gt;
wantsMoreyes  -0.8330     0.1175  -7.091 1.33e-12 ***&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
In a ''Generalized Linear Model (GLM)'' with a ''binomial family'' and a ''logit link'' (the default for ''R''), the coefficients represent '''log-odds ratios'''.&lt;br /&gt;
Since our outcome is structured as &lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
   cbind(using, notUsing),&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
we are modeling the probability of &amp;quot;using&amp;quot; (success) versus &amp;quot;not using&amp;quot; (failure).&lt;br /&gt;
&lt;br /&gt;
That is, the model interpretation is in terms of '''Log-Odds'''.&lt;br /&gt;
In this model, the relationship between the ''covariate predictors'' and the ''outcome'' is defined by the ''logit'' function&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\text{logit}(p) = \ln\left(\frac{p}{1-p}\right) = \beta_0 + \beta_1X_1 + \dots + \beta_kX_k.&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The estimate ''Sign'' matters:&lt;br /&gt;
&lt;br /&gt;
: ''Positive Estimate'': As the predictor increases (or if the category is present), the probability of &amp;quot;using&amp;quot; increases.&lt;br /&gt;
: ''Negative Estimate'': As the predictor increases, the probability of &amp;quot;using&amp;quot; decreases.&lt;br /&gt;
&lt;br /&gt;
The estimate ''Magnitude'' is interpreted via the &amp;quot;Exponentiation Trick&amp;quot;.&lt;br /&gt;
Because humans don't think naturally in &amp;quot;log-odds&amp;quot; terms, we usually ''exponentiate'' the coefficients (&amp;lt;math&amp;gt;e^\beta&amp;lt;/math&amp;gt;) to get the ''Odds Ratios (OR)''.&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Variable&lt;br /&gt;
! Estimate (β)&lt;br /&gt;
! Odds Ratio (e&amp;lt;sup&amp;gt;β&amp;lt;/sup&amp;gt;)&lt;br /&gt;
! Interpretation&lt;br /&gt;
|-&lt;br /&gt;
| '''age40-49'''&lt;br /&gt;
| 1.1892&lt;br /&gt;
| ≈ 3.28&lt;br /&gt;
| Women aged 40-49 have '''3.28 times the odds''' of using contraception compared to the reference age group (likely &amp;lt;25).&lt;br /&gt;
|-&lt;br /&gt;
| '''wantsMoreyes'''&lt;br /&gt;
| -0.8330&lt;br /&gt;
| ≈ 0.43&lt;br /&gt;
| Women who want more children have '''57% lower odds''' (1 - 0.43) of using contraception than those who don't.&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
Specific Breakdown of the Results:&lt;br /&gt;
&lt;br /&gt;
* ''Age'' (Categorical): R has treated age as a ''factor''. The baseline (reference) group is the youngest group (''under 25'').&lt;br /&gt;
:: As age increases (&amp;lt;math&amp;gt;25-29 \ \to 30-39\  \to 40-49&amp;lt;/math&amp;gt;), the coefficients become increasingly positive (&amp;lt;math&amp;gt;0.38 \to 0.90 \to 1.18&amp;lt;/math&amp;gt;).&lt;br /&gt;
:: Meaning: Older women in this dataset are significantly more likely to use contraception than the youngest group.&lt;br /&gt;
&lt;br /&gt;
* Education:&lt;br /&gt;
:: Estimate: -0.3250&lt;br /&gt;
:: Meaning: Being in the ''low education'' group is associated with a decrease in the log-odds of using contraception compared to the ''high'' group. Specifically, their odds are about 28\% lower (&amp;lt;math&amp;gt;e^{-0.325} \approx 0.72&amp;lt;/math&amp;gt;).&lt;br /&gt;
&lt;br /&gt;
* Wants More Children:&lt;br /&gt;
:: Estimate: -0.8330&lt;br /&gt;
:: Meaning: This is a strong negative predictor. If a woman wants more children, the probability of her using contraception drops significantly.&lt;br /&gt;
&lt;br /&gt;
We can have a quick &amp;quot;reality check&amp;quot; on the measuring units. Unlike a standard correlation (which is bounded between -1 and 1), these estimates can be any real number.&lt;br /&gt;
&lt;br /&gt;
:: An estimate of 0 means the variable has no effect on the odds (Odds Ratio = 1).&lt;br /&gt;
:: The z-value tells us how many standard errors the estimate is from zero. Since all p-values are small (&amp;lt;math&amp;gt;&amp;lt; 0.05&amp;lt;/math&amp;gt;), all these predictors are statistically significant.&lt;br /&gt;
&lt;br /&gt;
====Example 2: Poisson Regression for Count Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with built-in R dataset: AIDS cases in Belgium&lt;br /&gt;
# Load necessary libraries&lt;br /&gt;
if (!require(&amp;quot;MASS&amp;quot;)) install.packages(&amp;quot;MASS&amp;quot;)&lt;br /&gt;
library(MASS)&lt;br /&gt;
&lt;br /&gt;
# Load AIDS dataset&lt;br /&gt;
data(Aids2)&lt;br /&gt;
cat(&amp;quot;AIDS dataset structure:\n&amp;quot;)&lt;br /&gt;
str(Aids2)&lt;br /&gt;
&lt;br /&gt;
# Prepare data: count of AIDS cases by year and state&lt;br /&gt;
aids_counts &amp;lt;- aggregate(cbind(count = 1:nrow(Aids2)) ~ age + state,&lt;br /&gt;
                         data = Aids2, FUN = length)&lt;br /&gt;
&lt;br /&gt;
# Fit Poisson regression&lt;br /&gt;
model2 &amp;lt;- glm(count ~ age + state, family = poisson, data = aids_counts)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Poisson Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model2)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Check for Overdispersion ===\n&amp;quot;)&lt;br /&gt;
# Pearson chi-square statistic&lt;br /&gt;
pearson_chi2 &amp;lt;- sum(residuals(model2, type = &amp;quot;pearson&amp;quot;)^2)&lt;br /&gt;
df_resid &amp;lt;- df.residual(model2)&lt;br /&gt;
dispersion &amp;lt;- pearson_chi2 / df_resid&lt;br /&gt;
cat(&amp;quot;Pearson χ²:&amp;quot;, pearson_chi2, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;Dispersion parameter:&amp;quot;, dispersion, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;p-value for H0: φ=1:&amp;quot;, pchisq(pearson_chi2, df_resid, lower.tail = FALSE), &amp;quot;\n&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# If overdispersed, fit quasipoisson model&lt;br /&gt;
if (dispersion &amp;gt; 1.5) {&lt;br /&gt;
  cat(&amp;quot;\n=== Fitting Quasi-Poisson Model (accounting for overdispersion) ===\n&amp;quot;)&lt;br /&gt;
  model2_qp &amp;lt;- glm(count ~ age + state, family = quasipoisson, data = aids_counts)&lt;br /&gt;
  summary(model2_qp)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Predict expected counts&lt;br /&gt;
cat(&amp;quot;\n=== Predictions for First 10 Observations ===\n&amp;quot;)&lt;br /&gt;
predictions &amp;lt;- predict(model2, type = &amp;quot;response&amp;quot;, se.fit = TRUE)&lt;br /&gt;
pred_df &amp;lt;- data.frame(&lt;br /&gt;
  Observed = aids_counts$count[1:10],&lt;br /&gt;
  Predicted = predictions$fit[1:10],&lt;br /&gt;
  SE = predictions$se.fit[1:10],&lt;br /&gt;
  Lower = predictions$fit[1:10] - 1.96 * predictions$se.fit[1:10],&lt;br /&gt;
  Upper = predictions$fit[1:10] + 1.96 * predictions$se.fit[1:10]&lt;br /&gt;
)&lt;br /&gt;
print(pred_df)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Example 3: Gamma Regression for Positive Continuous Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with insurance claims data&lt;br /&gt;
if (!require(&amp;quot;insuranceData&amp;quot;)) install.packages(&amp;quot;insuranceData&amp;quot;)&lt;br /&gt;
library(insuranceData)&lt;br /&gt;
data(dataCar)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;Car insurance claims dataset:\n&amp;quot;)&lt;br /&gt;
str(dataCar)&lt;br /&gt;
&lt;br /&gt;
# Filter for positive claim amounts&lt;br /&gt;
claims_positive &amp;lt;- subset(dataCar, claimcst0 &amp;gt; 0)&lt;br /&gt;
&lt;br /&gt;
# Fit Gamma GLM with log link (common for monetary amounts)&lt;br /&gt;
model3 &amp;lt;- glm(claimcst0 ~ agecat + area + veh_age,&lt;br /&gt;
              family = Gamma(link = &amp;quot;log&amp;quot;),&lt;br /&gt;
              data = claims_positive)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Gamma Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model3)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Checking Gamma Model Assumptions ===\n&amp;quot;)&lt;br /&gt;
# Check residuals&lt;br /&gt;
res_gamma &amp;lt;- residuals(model3, type = &amp;quot;deviance&amp;quot;)&lt;br /&gt;
par(mfrow = c(1, 2))&lt;br /&gt;
hist(res_gamma, main = &amp;quot;Deviance Residuals&amp;quot;, xlab = &amp;quot;Residuals&amp;quot;)&lt;br /&gt;
qqnorm(res_gamma, main = &amp;quot;Q-Q Plot of Residuals&amp;quot;)&lt;br /&gt;
qqline(res_gamma)&lt;br /&gt;
&lt;br /&gt;
# Scale-location plot&lt;br /&gt;
fitted_values &amp;lt;- fitted(model3)&lt;br /&gt;
plot(fitted_values, sqrt(abs(res_gamma)),&lt;br /&gt;
     xlab = &amp;quot;Fitted Values&amp;quot;, ylab = &amp;quot;√|Deviance Residuals|&amp;quot;,&lt;br /&gt;
     main = &amp;quot;Scale-Location Plot&amp;quot;)&lt;br /&gt;
lines(lowess(fitted_values, sqrt(abs(res_gamma))), col = &amp;quot;red&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Software Implementation in R===&lt;br /&gt;
&lt;br /&gt;
Note that `family` is a function (a &amp;quot;closure&amp;quot;), not a list or an object with a \(\$family\) component. In thecode, we're passing&lt;br /&gt;
the family function itself (e.g., gaussian) to the \(run\_glm\_analysis()\) function, and later accessing \(family\$family\), &lt;br /&gt;
which doesn’t exist—because gaussian is a function, not a fitted model object.&lt;br /&gt;
In R, gaussian, binomial, poisson, etc., are functions that return family objects when called.&lt;br /&gt;
We're passing the function, not the result of calling it.&lt;br /&gt;
However, `glm()` internally calls `family()`, i.e., `gaussian()`, to get the actual family list object, which does have a \(\$family\) element.&lt;br /&gt;
Instead of checking \(family\$family\), we extract the family name from the ''fitted model'', not from the input argument.&lt;br /&gt;
Specifically, after fitting the model, we use \(model\$family\$family\), which is a character string like &amp;quot;gaussian&amp;quot;, &amp;quot;binomial&amp;quot;, etc.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Comprehensive GLM analysis function&lt;br /&gt;
run_glm_analysis &amp;lt;- function(formula, data, family, link = NULL) {&lt;br /&gt;
  # Fit the model&lt;br /&gt;
  if (!is.null(link)) {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family(link = link))&lt;br /&gt;
  } else {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family)&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  # Model summary&lt;br /&gt;
  cat(&amp;quot;=== MODEL SUMMARY ===\n&amp;quot;)&lt;br /&gt;
  print(summary(model))&lt;br /&gt;
  &lt;br /&gt;
  # Confidence intervals&lt;br /&gt;
  cat(&amp;quot;\n=== 95% CONFIDENCE INTERVALS ===\n&amp;quot;)&lt;br /&gt;
  print(confint(model))&lt;br /&gt;
  &lt;br /&gt;
  # Goodness-of-fit tests&lt;br /&gt;
  cat(&amp;quot;\n=== GOODNESS-OF-FIT ===\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Null Deviance:&amp;quot;, model$null.deviance, &amp;quot;on&amp;quot;, model$df.null, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Residual Deviance:&amp;quot;, model$deviance, &amp;quot;on&amp;quot;, model$df.residual, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;AIC:&amp;quot;, AIC(model), &amp;quot;\n&amp;quot;)&lt;br /&gt;
  &lt;br /&gt;
  # Check for overdispersion (for Poisson and binomial families)&lt;br /&gt;
  family_name &amp;lt;- model$family$family  # ✅ Extract from fitted model&lt;br /&gt;
  if (family_name %in% c(&amp;quot;poisson&amp;quot;, &amp;quot;binomial&amp;quot;, &amp;quot;quasipoisson&amp;quot;, &amp;quot;quasibinomial&amp;quot;)) {&lt;br /&gt;
    cat(&amp;quot;\n=== OVERDISPERSION CHECK ===\n&amp;quot;)&lt;br /&gt;
    dispersion &amp;lt;- model$deviance / model$df.residual&lt;br /&gt;
    cat(&amp;quot;Dispersion parameter:&amp;quot;, round(dispersion, 4), &amp;quot;\n&amp;quot;)&lt;br /&gt;
    if (abs(dispersion - 1) &amp;gt; 0.1) {&lt;br /&gt;
      direction &amp;lt;- ifelse(dispersion &amp;gt; 1, &amp;quot;over&amp;quot;, &amp;quot;under&amp;quot;)&lt;br /&gt;
      cat(&amp;quot;Note: Significant&amp;quot;, direction, &amp;quot;dispersion detected\n&amp;quot;)&lt;br /&gt;
    }&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  return(model)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Example usage with mtcars dataset&lt;br /&gt;
cat(&amp;quot;\n\n=== EXAMPLE: GAUSSIAN GLM (equivalent to linear regression) ===\n&amp;quot;)&lt;br /&gt;
data(mtcars)&lt;br /&gt;
model_gaussian &amp;lt;- run_glm_analysis(&lt;br /&gt;
  formula = mpg ~ wt + hp + cyl,&lt;br /&gt;
  data = mtcars,&lt;br /&gt;
  family = gaussian,&lt;br /&gt;
  link = &amp;quot;identity&amp;quot;&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Compare with lm() for verification&lt;br /&gt;
cat(&amp;quot;\n=== COMPARISON WITH lm() ===\n&amp;quot;)&lt;br /&gt;
model_lm &amp;lt;- lm(mpg ~ wt + hp + cyl, data = mtcars)&lt;br /&gt;
print(summary(model_lm))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Common GLM Families and Links in R===&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Family !! Default Link !! Alternative Links !! Typical Use&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || Continuous, symmetric data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;binomial()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;logit&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;probit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cauchit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cloglog&amp;lt;/code&amp;gt; || Binary/count proportions&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;poisson()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;sqrt&amp;lt;/code&amp;gt; || Count data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;Gamma()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, right-skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;inverse.gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;1/μ²&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, highly skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;quasi()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || User-defined || Overdispersed data&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Practical Considerations===&lt;br /&gt;
&lt;br /&gt;
* Model Selection:&lt;br /&gt;
** AIC (Akaike Information Criterion): \( \text{AIC} = -2\ell + 2p \)&lt;br /&gt;
** BIC (Bayesian Information Criterion): \( \text{BIC} = -2\ell + p\log(n) \)&lt;br /&gt;
** Cross-validation: Particularly useful for predictive performance.&lt;br /&gt;
&lt;br /&gt;
* Handling Overdispersion: For Poisson models: \( \text{Var}(Y) = \phi\mu \) where \(\phi &amp;gt; 1\) indicates overdispersion&lt;br /&gt;
Solutions:&lt;br /&gt;
** Use quasi-Poisson model&lt;br /&gt;
** Use negative binomial distribution&lt;br /&gt;
** Include random effects.&lt;br /&gt;
&lt;br /&gt;
* Zero-Inflation: For count data with excess zeros, consider:&lt;br /&gt;
** Zero-inflated Poisson (ZIP) model&lt;br /&gt;
** Zero-inflated negative binomial (ZINB) model&lt;br /&gt;
** Hurdle models.&lt;br /&gt;
&lt;br /&gt;
===Advanced Topics===&lt;br /&gt;
&lt;br /&gt;
==== Mixed Effects GLM (GLMM)====&lt;br /&gt;
&lt;br /&gt;
Extension incorporating random effects:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
g(E[Y|b]) = \mathbf{X}\beta + \mathbf{Z}b&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \( b \sim N(0, \mathbf{G}) \).&lt;br /&gt;
&lt;br /&gt;
Implementation in R with &amp;lt;code&amp;gt;lme4::glmer()&amp;lt;/code&amp;gt;:&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(lme4)&lt;br /&gt;
&lt;br /&gt;
# generate random data&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
id &amp;lt;- seq(n)&lt;br /&gt;
day &amp;lt;- 1:20&lt;br /&gt;
mydata &amp;lt;- expand.grid(id = id, day = day)&lt;br /&gt;
set.seed(1)&lt;br /&gt;
trt &amp;lt;- sample(c(&amp;quot;control&amp;quot;, &amp;quot;treat&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
sex &amp;lt;- sample(c(&amp;quot;female&amp;quot;, &amp;quot;male&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
mydata$trt &amp;lt;- trt[mydata$id]&lt;br /&gt;
mydata$sex &amp;lt;- sex[mydata$id]&lt;br /&gt;
mydata &amp;lt;- mydata[order(mydata$id, mydata$day),]&lt;br /&gt;
rownames(mydata) &amp;lt;- NULL&lt;br /&gt;
head(mydata, n = 10)&lt;br /&gt;
&lt;br /&gt;
mydata$trtsex &amp;lt;- interaction(mydata$trt, mydata$sex)&lt;br /&gt;
probs &amp;lt;- c(0.40, 0.85, 0.30, 0.50)&lt;br /&gt;
names(probs) &amp;lt;- levels(mydata$trtsex)&lt;br /&gt;
mydata$p &amp;lt;- probs[mydata$trtsex]&lt;br /&gt;
&lt;br /&gt;
set.seed(3)&lt;br /&gt;
r_probs &amp;lt;- rnorm(n = n, mean = 0, sd = 0.03)&lt;br /&gt;
mydata$random_p &amp;lt;- r_probs[mydata$id]&lt;br /&gt;
mydata$p &amp;lt;- mydata$p + mydata$random_p&lt;br /&gt;
&lt;br /&gt;
# use the probabilities to generate zeroes and ones with the binom() function.&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n = nrow(mydata), size = 1, prob = mydata$p)&lt;br /&gt;
&lt;br /&gt;
# using the sim data, inspect the first few records.&lt;br /&gt;
&lt;br /&gt;
head(mydata[c(&amp;quot;id&amp;quot;, &amp;quot;day&amp;quot;, &amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;, &amp;quot;p&amp;quot;, &amp;quot;y&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
# Example with binary outcome and random intercept&lt;br /&gt;
m &amp;lt;- glmer(y ~ trt * sex + (1|id), data = mydata, family = binomial)&lt;br /&gt;
summary(m, corr = FALSE)&lt;br /&gt;
&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Bayesian GLM====&lt;br /&gt;
&lt;br /&gt;
Using Markov Chain Monte Carlo (MCMC) methods.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
# Bayesian logistic regression&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(y ~ day +  trt + sex + p, # predictors,&lt;br /&gt;
                        family = binomial,&lt;br /&gt;
                        data = mydata,&lt;br /&gt;
                        prior = normal(0, 2.5),&lt;br /&gt;
                        prior_intercept = normal(0, 5))&lt;br /&gt;
&lt;br /&gt;
print(bayes_model, digits = 2)&lt;br /&gt;
&lt;br /&gt;
# Coefficient Plot (Forest Plot)&lt;br /&gt;
&lt;br /&gt;
plot(bayes_model, &amp;quot;areas&amp;quot;)  # density + intervals&lt;br /&gt;
# OR more customizable:&lt;br /&gt;
library(bayesplot)&lt;br /&gt;
mcmc_intervals(bayes_model, prob = 0.95) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Distributions of Coefficients&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Posterior Predictive Checks (PPC)&lt;br /&gt;
# Check if the model can reproduce data like the observed&lt;br /&gt;
pp_check(bayes_model, plotfun = &amp;quot;hist&amp;quot;, nreps = 100) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Predictive Check (Histogram of y_rep vs y)&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Another Bayesian Experiment.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
set.seed(123)&lt;br /&gt;
&lt;br /&gt;
# Simulate data — explicitly make trt and sex into factors&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
mydata &amp;lt;- data.frame(&lt;br /&gt;
  day = sample(1:30, n, replace = TRUE),&lt;br /&gt;
  trt = factor(sample(c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;)),&lt;br /&gt;
  sex = factor(sample(c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;)),&lt;br /&gt;
  p   = rnorm(n, mean = 50, sd = 10)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# True log-odds&lt;br /&gt;
logit_p &amp;lt;- -1 - 0.02 * mydata$day + &lt;br /&gt;
           0.6 * (as.numeric(mydata$trt) - 1) +   # A=0, B=1&lt;br /&gt;
           0.3 * (as.numeric(mydata$sex) - 1) +    # F=0, M=1&lt;br /&gt;
           0.04 * mydata$p&lt;br /&gt;
&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n, size = 1, prob = plogis(logit_p))&lt;br /&gt;
&lt;br /&gt;
# Fit model&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(&lt;br /&gt;
  y ~ day + trt + sex + p,&lt;br /&gt;
  family = binomial,&lt;br /&gt;
  data = mydata,&lt;br /&gt;
  prior = normal(0, 2.5),&lt;br /&gt;
  prior_intercept = normal(0, 5),&lt;br /&gt;
  seed = 42,&lt;br /&gt;
  refresh = 0&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Create New Data&lt;br /&gt;
# Confirm variables are factors&lt;br /&gt;
str(mydata[c(&amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
# Create prediction grid&lt;br /&gt;
newdata &amp;lt;- expand.grid(&lt;br /&gt;
  day = seq(min(mydata$day), max(mydata$day), length.out = 30),&lt;br /&gt;
  trt = levels(mydata$trt)[1],      # e.g., &amp;quot;A&amp;quot;&lt;br /&gt;
  sex = levels(mydata$sex)[1],      # e.g., &amp;quot;F&amp;quot;&lt;br /&gt;
  p   = median(mydata$p)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Verify it worked&lt;br /&gt;
stopifnot(nrow(newdata) &amp;gt; 0)&lt;br /&gt;
str(newdata)&lt;br /&gt;
&lt;br /&gt;
# Generate Predictions from the Bayesian Posterior Probability (after fitting the Bayesian Model):&lt;br /&gt;
# Posterior predicted probabilities&lt;br /&gt;
post_pred &amp;lt;- posterior_linpred(bayes_model, newdata = newdata, transform = TRUE)&lt;br /&gt;
&lt;br /&gt;
# Summarize&lt;br /&gt;
pred_summary &amp;lt;- data.frame(&lt;br /&gt;
  day = newdata$day,&lt;br /&gt;
  prob = apply(post_pred, 2, median),&lt;br /&gt;
  lower = apply(post_pred, 2, quantile, 0.025),&lt;br /&gt;
  upper = apply(post_pred, 2, quantile, 0.975)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Plot&lt;br /&gt;
library(ggplot2)&lt;br /&gt;
ggplot(pred_summary, aes(x = day, y = prob)) +&lt;br /&gt;
  geom_line(color = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  geom_ribbon(aes(ymin = lower, ymax = upper), alpha = 0.2, fill = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  labs(&lt;br /&gt;
    title = &amp;quot;Predicted Probability of Outcome by Day&amp;quot;,&lt;br /&gt;
    subtitle = &amp;quot;Holding treatment = A, sex = F, and p = median&amp;quot;,&lt;br /&gt;
    y = &amp;quot;P(y = 1)&amp;quot;,&lt;br /&gt;
    x = &amp;quot;Day&amp;quot;&lt;br /&gt;
  ) +&lt;br /&gt;
  theme_minimal()&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
==== ''Gamma GLM'' vs. ''Beta Regression''====&lt;br /&gt;
&lt;br /&gt;
The ''Gamma GLM'' and ''Beta Regression'' models are designed for very different types of &lt;br /&gt;
&amp;quot;continuous&amp;quot; data.&lt;br /&gt;
The fundamental difference lies in the '''domain''' (the range of possible values) and the &lt;br /&gt;
'''variance structure''' of the data they expect.&lt;br /&gt;
&lt;br /&gt;
* The Gamma Model (''glm''): The Gamma distribution is used for ''strictly positive, continuous data'' &amp;lt;math&amp;gt;(0, \infty)&amp;lt;/math&amp;gt;. It is appropriate for data that is &amp;quot;skewed&amp;quot; to the right, where the variance increases as the mean increases.&lt;br /&gt;
If a variable, like ''qol_score'' is a raw score (e.g., 0 to 100), the link, ''link = &amp;quot;log&amp;quot;'',&lt;br /&gt;
ensures the predicted values remain positive.&lt;br /&gt;
&lt;br /&gt;
In a Gamma GLM with a log link, we model the mean &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt; as&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\ln(\mu_i) = \beta_0 + \beta_1 X_1 + \beta_2 X_2 + \dots + \beta_k X_k.&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The probability density function (PDF) for the Gamma distribution is&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;f(y; \alpha, \beta) = \frac{\beta^\alpha y^{\alpha-1} e^{-\beta y}}{\Gamma(\alpha)} \quad \text{for } y &amp;gt; 0.&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
* The Beta Model (''betareg''), with Beta distribution, is used specifically for '''rates, proportions, or scores constrained between 0 and 1''', &amp;lt;math&amp;gt;(0, 1)&amp;lt;/math&amp;gt;. Unlike the Gamma model, which can go off to infinity, the Beta distribution is &amp;quot;boxed in&amp;quot; this interval.&lt;br /&gt;
For instance, if ''qol_scaled'' is normalized, e.g., by dividing QoL score by its maximum value,&lt;br /&gt;
it's values are forced into &amp;lt;math&amp;gt;(0, 1)&amp;lt;/math&amp;gt; interval.&lt;br /&gt;
Beta distribution shape is veryflexible. It can be U-shaped, J-shaped, or symmetric, making it perfect for &amp;quot;ceiling&amp;quot; or &amp;quot;floor&amp;quot; effects in Quality of Life scores.&lt;br /&gt;
&lt;br /&gt;
Beta regression typically uses a '''logit link''' by default to keep predictions between 0 and 1:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\ln\left(\frac{\mu_i}{1-\mu_i}\right) = \beta_0 + \beta_1 X_1 + \dots + \beta_k X_k.&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The PDF for the Beta distribution (parameterized by mean &amp;lt;math&amp;gt;\mu&amp;lt;/math&amp;gt; and precision &amp;lt;math&amp;gt;\phi&amp;lt;/math&amp;gt;) is:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;f(y; \mu, \phi) = \frac{\Gamma(\phi)}{\Gamma(\mu\phi)\Gamma((1-\mu)\phi)} y^{\mu\phi-1} (1-y)^{(1-\mu)\phi-1} \quad \text{for } y \in (0, 1).&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Key Differences between ''Gamma GLM'' and ''Beta Regression''&lt;br /&gt;
&lt;br /&gt;
| Feature | Gamma Model (model4) | Beta Model (`model_beta`) |&lt;br /&gt;
| --- | --- | --- |&lt;br /&gt;
| Outcome Range | &amp;lt;math&amp;gt;(0, \infty)&amp;lt;/math&amp;gt; (Positive numbers) | &amp;lt;math&amp;gt;(0, 1)&amp;lt;/math&amp;gt; (Proportions/Scaled) |&lt;br /&gt;
| Typical Use | Costs, rainfall, raw test scores | Percentages, scaled QoL indices |&lt;br /&gt;
| Interpretation | Log-link: Coefficients are multiplicative | Logit-link: Coefficients are odds ratios |&lt;br /&gt;
| Boundaries | Can handle very large values | Naturally handles &amp;quot;ceiling effects&amp;quot; |&lt;br /&gt;
&lt;br /&gt;
If the outcome dependent variable (DV) qol_score is a raw sum that could theoretically be much higher, stick with Gamma. But when ''qol_score'' is a bounded instrument (like a visual analog scale from 0–1) with many observations scoring near the very top or bottom, ''Beta'' may be superior because it understands the &amp;quot;walls&amp;quot; of the scale.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(readr)&lt;br /&gt;
 &lt;br /&gt;
 # REMOTE URL&lt;br /&gt;
 url_wide &amp;lt;- &amp;quot;https://socr.umich.edu/docs/uploads/2026/SOCR_CRDS_CompAnalysis_CaseStudy_wide.csv&amp;quot;&lt;br /&gt;
 url_long &amp;lt;- &amp;quot;https://socr.umich.edu/docs/uploads/2026/SOCR_CRDS_CompAnalysis_CaseStudy_long.csv&amp;quot;&lt;br /&gt;
 &lt;br /&gt;
 # Import: Reading the Wide data&lt;br /&gt;
 df_wide &amp;lt;- read_csv(url_wide, col_types = cols(&lt;br /&gt;
     patient_id      = col_factor(),    # Prevents math on IDs&lt;br /&gt;
     age             = col_double(),&lt;br /&gt;
     gender          = col_character(),&lt;br /&gt;
     treatment       = col_character(),&lt;br /&gt;
     smoking_history = col_character(),&lt;br /&gt;
     baseline_fvc    = col_double(),&lt;br /&gt;
     fvc_0           = col_double(),&lt;br /&gt;
     fvc_6           = col_double(),&lt;br /&gt;
     fvc_12          = col_double(),&lt;br /&gt;
     qol_score       = col_double(),&lt;br /&gt;
     walk_dist       = col_double(),&lt;br /&gt;
     success         = col_integer(),   # 0 or 1 is best as integer&lt;br /&gt;
     hospital_visits = col_integer(),&lt;br /&gt;
     physician_notes = col_character()&lt;br /&gt;
 ))&lt;br /&gt;
                                                                                   &lt;br /&gt;
 # Import: Reading the Long data&lt;br /&gt;
 df_long &amp;lt;- read_csv(url_long, col_types = cols(&lt;br /&gt;
     patient_id = col_factor(),  &lt;br /&gt;
     age = col_double(),&lt;br /&gt;
     gender = col_character(),&lt;br /&gt;
     timepoint = col_factor(levels = c(&amp;quot;Month 0&amp;quot;, &amp;quot;Month 6&amp;quot;, &amp;quot;Month 12&amp;quot;))&lt;br /&gt;
 ))&lt;br /&gt;
&lt;br /&gt;
df_wide$qol_scaled &amp;lt;- df_wide$qol_score / (max(df_wide$qol_score) + 0.0001) &lt;br /&gt;
 &lt;br /&gt;
 # Use the betareg package&lt;br /&gt;
 library(betareg)&lt;br /&gt;
&lt;br /&gt;
model4 &amp;lt;- glm(qol_score ~ treatment + gender + age + baseline_fvc + success, family = Gamma(link = &amp;quot;log&amp;quot;), data = df_wide)&lt;br /&gt;
summary(model4)&lt;br /&gt;
model_beta &amp;lt;- betareg(qol_scaled ~ treatment + gender + baseline_fvc + success + age, data = df_wide)&lt;br /&gt;
summary(model_beta )&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Problems===&lt;br /&gt;
&lt;br /&gt;
1. Conceptual Exercises:&lt;br /&gt;
&lt;br /&gt;
   a) Derive the score equations for a Poisson GLM with log link&lt;br /&gt;
   b) Show that the binomial distribution with logit link is the canonical link&lt;br /&gt;
   c) Prove that the deviance for a normal GLM equals the residual sum of squares&lt;br /&gt;
&lt;br /&gt;
2. Applied Problems:&lt;br /&gt;
&lt;br /&gt;
   a) Analyze the [http://wiki.socr.umich.edu/index.php/SOCR_Data_Dinov_021808_ConsumerPriceIndex3Way Consumer Price Index] data using appropriate GLM&lt;br /&gt;
   b) Model the [http://wiki.socr.umich.edu/index.php/SOCR_Data_MonetaryBase1959_2009 Monetary Base] data considering temporal autocorrelation&lt;br /&gt;
   c) Using the &amp;lt;code&amp;gt;iris&amp;lt;/code&amp;gt; dataset, build a multinomial logistic regression to classify species&lt;br /&gt;
&lt;br /&gt;
3. Practice Simulation Study:&lt;br /&gt;
&lt;br /&gt;
   &amp;lt;pre&amp;gt;&lt;br /&gt;
   # Simulate data from a logistic regression model&lt;br /&gt;
   set.seed(123)&lt;br /&gt;
   n &amp;lt;- 1000&lt;br /&gt;
   x1 &amp;lt;- rnorm(n)&lt;br /&gt;
   x2 &amp;lt;- rnorm(n)&lt;br /&gt;
   beta &amp;lt;- c(0.5, 1, -0.5)&lt;br /&gt;
   linear_predictor &amp;lt;- beta[1] + beta[2]*x1 + beta[3]*x2&lt;br /&gt;
   probabilities &amp;lt;- plogis(linear_predictor)&lt;br /&gt;
   y &amp;lt;- rbinom(n, size = 1, prob = probabilities)&lt;br /&gt;
   &lt;br /&gt;
   # Fit model and evaluate performance&lt;br /&gt;
   sim_data &amp;lt;- data.frame(y = y, x1 = x1, x2 = x2)&lt;br /&gt;
   model_sim &amp;lt;- glm(y ~ x1 + x2, family = binomial, data = sim_data)&lt;br /&gt;
   &lt;br /&gt;
   # Calculate bias and MSE&lt;br /&gt;
   beta_hat &amp;lt;- coef(model_sim)&lt;br /&gt;
   bias &amp;lt;- beta_hat - beta&lt;br /&gt;
   mse &amp;lt;- mean((beta_hat - beta)^2)&lt;br /&gt;
   &lt;br /&gt;
   cat(&amp;quot;True parameters:&amp;quot;, beta, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Estimated parameters:&amp;quot;, beta_hat, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Bias:&amp;quot;, bias, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;MSE:&amp;quot;, mse, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   &amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
&lt;br /&gt;
1. McCullagh, P., &amp;amp; Nelder, J. A. (1989). *Generalized Linear Models* (2nd ed.). Chapman and Hall.&lt;br /&gt;
&lt;br /&gt;
2. Dobson, A. J., &amp;amp; Barnett, A. G. (2018). *An Introduction to Generalized Linear Models* (4th ed.). CRC Press.&lt;br /&gt;
&lt;br /&gt;
3. Agresti, A. (2015). *Foundations of Linear and Generalized Linear Models*. Wiley.&lt;br /&gt;
&lt;br /&gt;
4. Fahrmeir, L., Kneib, T., Lang, S., &amp;amp; Marx, B. (2013). *Regression: Models, Methods and Applications*. Springer.&lt;br /&gt;
&lt;br /&gt;
5. Wood, S. N. (2017). *Generalized Additive Models: An Introduction with R* (2nd ed.). Chapman and Hall/CRC.&lt;br /&gt;
&lt;br /&gt;
====Online Resources====&lt;br /&gt;
* [https://en.wikipedia.org/wiki/Generalized_linear_model GLM Wikipedia]&lt;br /&gt;
* [https://data.princeton.edu/wws509/notes/a2.pdf GLM Theory Notes]&lt;br /&gt;
* [https://www.jstatsoft.org/article/view/v015i12 GLM in R Tutorial]&lt;br /&gt;
* [https://cran.r-project.org/web/views/SocialSciences.html R Resources for Social Sciences]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php?title=SMHS_GLM}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18353</id>
		<title>SMHS GLM</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18353"/>
		<updated>2026-03-10T17:01:50Z</updated>

		<summary type="html">&lt;p&gt;Dinov: &lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Generalized Linear Modeling (GLM) ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Generalized Linear Modeling (GLM) is a flexible generalization of ordinary linear regression that allows response variables to have error distribution models other than a normal distribution. GLM extends linear regression by allowing the linear model to be related to the response variable via a link function and enabling the variance of each measurement to be a function of its predicted value. This framework unifies statistical models including linear regression, logistic regression, and Poisson regression. Estimation methods include iteratively reweighted least squares for maximum likelihood estimation, Bayesian approaches, and least squares fitted to variance-stabilized responses.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
While linear regression models linear relationships between response and predictors, many real-world scenarios involve response variables that don't follow normal distributions. For example:&lt;br /&gt;
* Binary outcomes (yes/no decisions) with probabilities bounded between 0 and 1&lt;br /&gt;
* Count data (number of events) that follow Poisson distributions&lt;br /&gt;
* Survival times that follow exponential or Weibull distributions&lt;br /&gt;
&lt;br /&gt;
GLM provides a unified framework for these situations by allowing response variables from the exponential family of distributions.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====1) GLM Components====&lt;br /&gt;
A GLM consists of three components:&lt;br /&gt;
&lt;br /&gt;
1. Random Component: The response variable \(Y\) follows a distribution from the exponential family:&lt;br /&gt;
   &amp;lt;math&amp;gt;f_Y(y|\theta,\phi) = \exp\left\{\frac{y\theta - b(\theta)}{a(\phi)} + c(y,\phi)\right\}&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;\theta&amp;lt;/math&amp;gt; is the natural parameter and \(\phi\) is the dispersion parameter.&lt;br /&gt;
&lt;br /&gt;
2. Systematic Component: The linear predictor \(\eta\):&lt;br /&gt;
   &amp;lt;math&amp;gt;\eta = \mathbf{X}\beta = \beta_0 + \beta_1X_1 + \beta_2X_2 + \cdots + \beta_pX_p&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
3. Link Function: \(g(\cdot)\) that relates the mean \(\mu = E[Y]\) to the linear predictor:&lt;br /&gt;
   &amp;lt;math&amp;gt;g(\mu) = \eta \quad \text{or equivalently} \quad \mu = g^{-1}(\eta)&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The variance function relates the variance to the mean: &amp;lt;math&amp;gt;\text{Var}(Y) = a(\phi)V(\mu)&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;V(\mu)&amp;lt;/math&amp;gt; is the variance function specific to the distribution.&lt;br /&gt;
&lt;br /&gt;
====2) Exponential Family Distributions====&lt;br /&gt;
The exponential family includes many common distributions:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Distribution !! Support !! Natural Parameter \(\theta\) !! \(b(\theta)\) !! Canonical Link \(g(\mu)\) !! Variance Function \(V(\mu)\)&lt;br /&gt;
|-&lt;br /&gt;
| Normal || \(\mathbb{R}\) || \(\mu\) || \(\frac{\theta^2}{2}\) || Identity: \(\mu\) || 1&lt;br /&gt;
|-&lt;br /&gt;
| Binomial || \(\{0,1,\ldots,n\}\) || \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(n\log(1+e^\theta)\) || Logit: \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(\mu\left(1-\frac{\mu}{n}\right)\)&lt;br /&gt;
|-&lt;br /&gt;
| Poisson || \(\mathbb{N}_0\) || \(\log(\mu)\) || \(e^\theta\) || Log: \(\log(\mu)\) || \(\mu\)&lt;br /&gt;
|-&lt;br /&gt;
| Gamma || \(\mathbb{R}^+\) || \(-\frac{1}{\mu}\) || \(-\log(-\theta)\) || Inverse: \(\frac{1}{\mu}\) || \(\mu^2\)&lt;br /&gt;
|-&lt;br /&gt;
| Inverse Gaussian || \(\mathbb{R}^+\) || \(-\frac{1}{2\mu^2}\) || \(-\sqrt{-2\theta}\) || Inverse squared: \(\frac{1}{\mu^2}\) || \(\mu^3\)&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====3) Maximum Likelihood Estimation====&lt;br /&gt;
For a GLM with \(n\) independent observations, the log-likelihood is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\ell(\beta) = \sum_{i=1}^n \frac{y_i\theta_i - b(\theta_i)}{a(\phi)} + c(y_i,\phi)&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\theta_i = \theta(\mu_i)\) and \(\mu_i = g^{-1}(\mathbf{x}_i^\top\beta)\).&lt;br /&gt;
&lt;br /&gt;
The score equations are:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathbf{U}(\beta) = \frac{\partial\ell}{\partial\beta} = \mathbf{X}^\top\mathbf{W}(\mathbf{y} - \boldsymbol{\mu}) = 0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\mathbf{W} = \text{diag}\left\{\frac{1}{a(\phi)V(\mu_i)[g'(\mu_i)]^2}\right\}\).&lt;br /&gt;
&lt;br /&gt;
The Fisher information matrix is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathcal{I}(\beta) = E\left[-\frac{\partial^2\ell}{\partial\beta\partial\beta^\top}\right] = \mathbf{X}^\top\mathbf{W}\mathbf{X}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Parameters are estimated via Iteratively Reweighted Least Squares (IRLS):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\beta^{(t+1)} = \beta^{(t)} + (\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{z}^{(t)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(z_i^{(t)} = \eta_i^{(t)} + (y_i - \mu_i^{(t)})g'(\mu_i^{(t)})\).&lt;br /&gt;
&lt;br /&gt;
====4) Deviance and Goodness-of-Fit====&lt;br /&gt;
The deviance measures goodness-of-fit:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D = 2[\ell(\text{saturated model}) - \ell(\text{fitted model})]&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
For nested models \(M_0 \subset M_1\):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D_{M_0} - D_{M_1} \sim \chi^2_{df_{M_0} - df_{M_1}} \quad \text{under } H_0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The scaled deviance is \(D^* = D/\phi\), and Pearson's chi-square statistic is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\chi^2_P = \sum_{i=1}^n \frac{(y_i - \hat{\mu}_i)^2}{V(\hat{\mu}_i)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====5) Model Diagnostics====&lt;br /&gt;
Key diagnostic tools:&lt;br /&gt;
* Pearson residuals: \(r_i^P = \frac{y_i - \hat{\mu}_i}{\sqrt{V(\hat{\mu}_i)}}\)&lt;br /&gt;
* Deviance residuals: \(r_i^D = \text{sign}(y_i - \hat{\mu}_i)\sqrt{d_i}\)&lt;br /&gt;
* Leverage: \(h_{ii}\) from the hat matrix \(\mathbf{H} = \mathbf{W}^{1/2}\mathbf{X}(\mathbf{X}^\top\mathbf{W}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{1/2}\)&lt;br /&gt;
* Cook's distance: \(D_i = \frac{r_i^2 h_{ii}}{p(1-h_{ii})}\)&lt;br /&gt;
&lt;br /&gt;
===Applications===&lt;br /&gt;
&lt;br /&gt;
====Example 1: Logistic Regression for Contraceptive Use====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Load and prepare data&lt;br /&gt;
# https://grodri.github.io/datasets/cuse.dat&lt;br /&gt;
cuse &amp;lt;- read.table(&amp;quot;https://grodri.github.io/datasets/cuse.dat&amp;quot;, header=TRUE)&lt;br /&gt;
cat(&amp;quot;First few rows of dataset:\n&amp;quot;)&lt;br /&gt;
print(head(cuse))&lt;br /&gt;
&lt;br /&gt;
# Fit binomial GLM with logit link&lt;br /&gt;
model1 &amp;lt;- glm(cbind(using, notUsing) ~ age + education + wantsMore,&lt;br /&gt;
              family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Confidence Intervals ===\n&amp;quot;)&lt;br /&gt;
confint(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Odds Ratios with 95% CI ===\n&amp;quot;)&lt;br /&gt;
exp_coef &amp;lt;- exp(coef(model1))&lt;br /&gt;
exp_ci &amp;lt;- exp(confint(model1))&lt;br /&gt;
cbind(OR = exp_coef, exp_ci)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Comparison (Likelihood Ratio Test) ===\n&amp;quot;)&lt;br /&gt;
# Reduced model without education&lt;br /&gt;
model_reduced &amp;lt;- glm(cbind(using, notUsing) ~ age + wantsMore,&lt;br /&gt;
                     family = binomial, data = cuse)&lt;br /&gt;
anova(model_reduced, model1, test = &amp;quot;Chisq&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Diagnostic plots&lt;br /&gt;
par(mfrow = c(2, 2))&lt;br /&gt;
plot(model1, which = 1:4)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
'''Result Interpretation''':&lt;br /&gt;
&lt;br /&gt;
Let's explicate ''how to interpret the parameter estimates in this GLM model''.&lt;br /&gt;
Specifically, as this is a bivariate outcome, the estimates are not correlations&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
glm(formula = cbind(using, notUsing) ~ age + education + wantsMore, &lt;br /&gt;
    family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
Coefficients:&lt;br /&gt;
             Estimate Std. Error z value Pr(&amp;gt;|z|)    &lt;br /&gt;
(Intercept)   -0.8082     0.1590  -5.083 3.71e-07 ***&lt;br /&gt;
age25-29       0.3894     0.1759   2.214  0.02681 *  &lt;br /&gt;
age30-39       0.9086     0.1646   5.519 3.40e-08 ***&lt;br /&gt;
age40-49       1.1892     0.2144   5.546 2.92e-08 ***&lt;br /&gt;
educationlow  -0.3250     0.1240  -2.620  0.00879 ** &lt;br /&gt;
wantsMoreyes  -0.8330     0.1175  -7.091 1.33e-12 ***&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
In a ''Generalized Linear Model (GLM)'' with a ''binomial family'' and a ''logit link'' (the default for ''R''), the coefficients represent '''log-odds ratios'''.&lt;br /&gt;
Since our outcome is structured as &lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
   cbind(using, notUsing),&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
we are modeling the probability of &amp;quot;using&amp;quot; (success) versus &amp;quot;not using&amp;quot; (failure).&lt;br /&gt;
&lt;br /&gt;
That is, the model interpretation is in terms of '''Log-Odds'''.&lt;br /&gt;
In this model, the relationship between the ''covariate predictors'' and the ''outcome'' is defined by the ''logit'' function&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\text{logit}(p) = \ln\left(\frac{p}{1-p}\right) = \beta_0 + \beta_1X_1 + \dots + \beta_kX_k.&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The estimate ''Sign'' matters:&lt;br /&gt;
&lt;br /&gt;
: ''Positive Estimate'': As the predictor increases (or if the category is present), the probability of &amp;quot;using&amp;quot; increases.&lt;br /&gt;
: ''Negative Estimate'': As the predictor increases, the probability of &amp;quot;using&amp;quot; decreases.&lt;br /&gt;
&lt;br /&gt;
The estimate ''Magnitude'' is interpreted via the &amp;quot;Exponentiation Trick&amp;quot;.&lt;br /&gt;
Because humans don't think naturally in &amp;quot;log-odds&amp;quot; terms, we usually ''exponentiate'' the coefficients (&amp;lt;math&amp;gt;e^\beta&amp;lt;/math&amp;gt;) to get the ''Odds Ratios (OR)''.&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Variable&lt;br /&gt;
! Estimate (β)&lt;br /&gt;
! Odds Ratio (e&amp;lt;sup&amp;gt;β&amp;lt;/sup&amp;gt;)&lt;br /&gt;
! Interpretation&lt;br /&gt;
|-&lt;br /&gt;
| '''age40-49'''&lt;br /&gt;
| 1.1892&lt;br /&gt;
| ≈ 3.28&lt;br /&gt;
| Women aged 40-49 have '''3.28 times the odds''' of using contraception compared to the reference age group (likely &amp;lt;25).&lt;br /&gt;
|-&lt;br /&gt;
| '''wantsMoreyes'''&lt;br /&gt;
| -0.8330&lt;br /&gt;
| ≈ 0.43&lt;br /&gt;
| Women who want more children have '''57% lower odds''' (1 - 0.43) of using contraception than those who don't.&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
Specific Breakdown of the Results:&lt;br /&gt;
&lt;br /&gt;
* ''Age'' (Categorical): R has treated age as a ''factor''. The baseline (reference) group is the youngest group (''under 25'').&lt;br /&gt;
:: As age increases (&amp;lt;math&amp;gt;25-29 \ \to 30-39\  \to 40-49&amp;lt;/math&amp;gt;), the coefficients become increasingly positive (&amp;lt;math&amp;gt;0.38 \to 0.90 \to 1.18&amp;lt;/math&amp;gt;).&lt;br /&gt;
:: Meaning: Older women in this dataset are significantly more likely to use contraception than the youngest group.&lt;br /&gt;
&lt;br /&gt;
* Education:&lt;br /&gt;
:: Estimate: -0.3250&lt;br /&gt;
:: Meaning: Being in the ''low education'' group is associated with a decrease in the log-odds of using contraception compared to the ''high'' group. Specifically, their odds are about 28\% lower (&amp;lt;math&amp;gt;e^{-0.325} \approx 0.72&amp;lt;/math&amp;gt;).&lt;br /&gt;
&lt;br /&gt;
* Wants More Children:&lt;br /&gt;
:: Estimate: -0.8330&lt;br /&gt;
:: Meaning: This is a strong negative predictor. If a woman wants more children, the probability of her using contraception drops significantly.&lt;br /&gt;
&lt;br /&gt;
We can have a quick &amp;quot;reality check&amp;quot; on the measuring units. Unlike a standard correlation (which is bounded between -1 and 1), these estimates can be any real number.&lt;br /&gt;
&lt;br /&gt;
:: An estimate of 0 means the variable has no effect on the odds (Odds Ratio = 1).&lt;br /&gt;
:: The z-value tells us how many standard errors the estimate is from zero. Since all p-values are small (&amp;lt;math&amp;gt;&amp;lt; 0.05&amp;lt;/math&amp;gt;), all these predictors are statistically significant.&lt;br /&gt;
&lt;br /&gt;
====Example 2: Poisson Regression for Count Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with built-in R dataset: AIDS cases in Belgium&lt;br /&gt;
# Load necessary libraries&lt;br /&gt;
if (!require(&amp;quot;MASS&amp;quot;)) install.packages(&amp;quot;MASS&amp;quot;)&lt;br /&gt;
library(MASS)&lt;br /&gt;
&lt;br /&gt;
# Load AIDS dataset&lt;br /&gt;
data(Aids2)&lt;br /&gt;
cat(&amp;quot;AIDS dataset structure:\n&amp;quot;)&lt;br /&gt;
str(Aids2)&lt;br /&gt;
&lt;br /&gt;
# Prepare data: count of AIDS cases by year and state&lt;br /&gt;
aids_counts &amp;lt;- aggregate(cbind(count = 1:nrow(Aids2)) ~ age + state,&lt;br /&gt;
                         data = Aids2, FUN = length)&lt;br /&gt;
&lt;br /&gt;
# Fit Poisson regression&lt;br /&gt;
model2 &amp;lt;- glm(count ~ age + state, family = poisson, data = aids_counts)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Poisson Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model2)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Check for Overdispersion ===\n&amp;quot;)&lt;br /&gt;
# Pearson chi-square statistic&lt;br /&gt;
pearson_chi2 &amp;lt;- sum(residuals(model2, type = &amp;quot;pearson&amp;quot;)^2)&lt;br /&gt;
df_resid &amp;lt;- df.residual(model2)&lt;br /&gt;
dispersion &amp;lt;- pearson_chi2 / df_resid&lt;br /&gt;
cat(&amp;quot;Pearson χ²:&amp;quot;, pearson_chi2, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;Dispersion parameter:&amp;quot;, dispersion, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;p-value for H0: φ=1:&amp;quot;, pchisq(pearson_chi2, df_resid, lower.tail = FALSE), &amp;quot;\n&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# If overdispersed, fit quasipoisson model&lt;br /&gt;
if (dispersion &amp;gt; 1.5) {&lt;br /&gt;
  cat(&amp;quot;\n=== Fitting Quasi-Poisson Model (accounting for overdispersion) ===\n&amp;quot;)&lt;br /&gt;
  model2_qp &amp;lt;- glm(count ~ age + state, family = quasipoisson, data = aids_counts)&lt;br /&gt;
  summary(model2_qp)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Predict expected counts&lt;br /&gt;
cat(&amp;quot;\n=== Predictions for First 10 Observations ===\n&amp;quot;)&lt;br /&gt;
predictions &amp;lt;- predict(model2, type = &amp;quot;response&amp;quot;, se.fit = TRUE)&lt;br /&gt;
pred_df &amp;lt;- data.frame(&lt;br /&gt;
  Observed = aids_counts$count[1:10],&lt;br /&gt;
  Predicted = predictions$fit[1:10],&lt;br /&gt;
  SE = predictions$se.fit[1:10],&lt;br /&gt;
  Lower = predictions$fit[1:10] - 1.96 * predictions$se.fit[1:10],&lt;br /&gt;
  Upper = predictions$fit[1:10] + 1.96 * predictions$se.fit[1:10]&lt;br /&gt;
)&lt;br /&gt;
print(pred_df)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Example 3: Gamma Regression for Positive Continuous Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with insurance claims data&lt;br /&gt;
if (!require(&amp;quot;insuranceData&amp;quot;)) install.packages(&amp;quot;insuranceData&amp;quot;)&lt;br /&gt;
library(insuranceData)&lt;br /&gt;
data(dataCar)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;Car insurance claims dataset:\n&amp;quot;)&lt;br /&gt;
str(dataCar)&lt;br /&gt;
&lt;br /&gt;
# Filter for positive claim amounts&lt;br /&gt;
claims_positive &amp;lt;- subset(dataCar, claimcst0 &amp;gt; 0)&lt;br /&gt;
&lt;br /&gt;
# Fit Gamma GLM with log link (common for monetary amounts)&lt;br /&gt;
model3 &amp;lt;- glm(claimcst0 ~ agecat + area + veh_age,&lt;br /&gt;
              family = Gamma(link = &amp;quot;log&amp;quot;),&lt;br /&gt;
              data = claims_positive)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Gamma Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model3)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Checking Gamma Model Assumptions ===\n&amp;quot;)&lt;br /&gt;
# Check residuals&lt;br /&gt;
res_gamma &amp;lt;- residuals(model3, type = &amp;quot;deviance&amp;quot;)&lt;br /&gt;
par(mfrow = c(1, 2))&lt;br /&gt;
hist(res_gamma, main = &amp;quot;Deviance Residuals&amp;quot;, xlab = &amp;quot;Residuals&amp;quot;)&lt;br /&gt;
qqnorm(res_gamma, main = &amp;quot;Q-Q Plot of Residuals&amp;quot;)&lt;br /&gt;
qqline(res_gamma)&lt;br /&gt;
&lt;br /&gt;
# Scale-location plot&lt;br /&gt;
fitted_values &amp;lt;- fitted(model3)&lt;br /&gt;
plot(fitted_values, sqrt(abs(res_gamma)),&lt;br /&gt;
     xlab = &amp;quot;Fitted Values&amp;quot;, ylab = &amp;quot;√|Deviance Residuals|&amp;quot;,&lt;br /&gt;
     main = &amp;quot;Scale-Location Plot&amp;quot;)&lt;br /&gt;
lines(lowess(fitted_values, sqrt(abs(res_gamma))), col = &amp;quot;red&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Software Implementation in R===&lt;br /&gt;
&lt;br /&gt;
Note that `family` is a function (a &amp;quot;closure&amp;quot;), not a list or an object with a \(\$family\) component. In thecode, we're passing&lt;br /&gt;
the family function itself (e.g., gaussian) to the \(run\_glm\_analysis()\) function, and later accessing \(family\$family\), &lt;br /&gt;
which doesn’t exist—because gaussian is a function, not a fitted model object.&lt;br /&gt;
In R, gaussian, binomial, poisson, etc., are functions that return family objects when called.&lt;br /&gt;
We're passing the function, not the result of calling it.&lt;br /&gt;
However, `glm()` internally calls `family()`, i.e., `gaussian()`, to get the actual family list object, which does have a \(\$family\) element.&lt;br /&gt;
Instead of checking \(family\$family\), we extract the family name from the ''fitted model'', not from the input argument.&lt;br /&gt;
Specifically, after fitting the model, we use \(model\$family\$family\), which is a character string like &amp;quot;gaussian&amp;quot;, &amp;quot;binomial&amp;quot;, etc.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Comprehensive GLM analysis function&lt;br /&gt;
run_glm_analysis &amp;lt;- function(formula, data, family, link = NULL) {&lt;br /&gt;
  # Fit the model&lt;br /&gt;
  if (!is.null(link)) {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family(link = link))&lt;br /&gt;
  } else {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family)&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  # Model summary&lt;br /&gt;
  cat(&amp;quot;=== MODEL SUMMARY ===\n&amp;quot;)&lt;br /&gt;
  print(summary(model))&lt;br /&gt;
  &lt;br /&gt;
  # Confidence intervals&lt;br /&gt;
  cat(&amp;quot;\n=== 95% CONFIDENCE INTERVALS ===\n&amp;quot;)&lt;br /&gt;
  print(confint(model))&lt;br /&gt;
  &lt;br /&gt;
  # Goodness-of-fit tests&lt;br /&gt;
  cat(&amp;quot;\n=== GOODNESS-OF-FIT ===\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Null Deviance:&amp;quot;, model$null.deviance, &amp;quot;on&amp;quot;, model$df.null, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Residual Deviance:&amp;quot;, model$deviance, &amp;quot;on&amp;quot;, model$df.residual, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;AIC:&amp;quot;, AIC(model), &amp;quot;\n&amp;quot;)&lt;br /&gt;
  &lt;br /&gt;
  # Check for overdispersion (for Poisson and binomial families)&lt;br /&gt;
  family_name &amp;lt;- model$family$family  # ✅ Extract from fitted model&lt;br /&gt;
  if (family_name %in% c(&amp;quot;poisson&amp;quot;, &amp;quot;binomial&amp;quot;, &amp;quot;quasipoisson&amp;quot;, &amp;quot;quasibinomial&amp;quot;)) {&lt;br /&gt;
    cat(&amp;quot;\n=== OVERDISPERSION CHECK ===\n&amp;quot;)&lt;br /&gt;
    dispersion &amp;lt;- model$deviance / model$df.residual&lt;br /&gt;
    cat(&amp;quot;Dispersion parameter:&amp;quot;, round(dispersion, 4), &amp;quot;\n&amp;quot;)&lt;br /&gt;
    if (abs(dispersion - 1) &amp;gt; 0.1) {&lt;br /&gt;
      direction &amp;lt;- ifelse(dispersion &amp;gt; 1, &amp;quot;over&amp;quot;, &amp;quot;under&amp;quot;)&lt;br /&gt;
      cat(&amp;quot;Note: Significant&amp;quot;, direction, &amp;quot;dispersion detected\n&amp;quot;)&lt;br /&gt;
    }&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  return(model)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Example usage with mtcars dataset&lt;br /&gt;
cat(&amp;quot;\n\n=== EXAMPLE: GAUSSIAN GLM (equivalent to linear regression) ===\n&amp;quot;)&lt;br /&gt;
data(mtcars)&lt;br /&gt;
model_gaussian &amp;lt;- run_glm_analysis(&lt;br /&gt;
  formula = mpg ~ wt + hp + cyl,&lt;br /&gt;
  data = mtcars,&lt;br /&gt;
  family = gaussian,&lt;br /&gt;
  link = &amp;quot;identity&amp;quot;&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Compare with lm() for verification&lt;br /&gt;
cat(&amp;quot;\n=== COMPARISON WITH lm() ===\n&amp;quot;)&lt;br /&gt;
model_lm &amp;lt;- lm(mpg ~ wt + hp + cyl, data = mtcars)&lt;br /&gt;
print(summary(model_lm))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Common GLM Families and Links in R===&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Family !! Default Link !! Alternative Links !! Typical Use&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || Continuous, symmetric data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;binomial()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;logit&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;probit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cauchit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cloglog&amp;lt;/code&amp;gt; || Binary/count proportions&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;poisson()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;sqrt&amp;lt;/code&amp;gt; || Count data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;Gamma()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, right-skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;inverse.gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;1/μ²&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, highly skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;quasi()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || User-defined || Overdispersed data&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Practical Considerations===&lt;br /&gt;
&lt;br /&gt;
* Model Selection:&lt;br /&gt;
** AIC (Akaike Information Criterion): \( \text{AIC} = -2\ell + 2p \)&lt;br /&gt;
** BIC (Bayesian Information Criterion): \( \text{BIC} = -2\ell + p\log(n) \)&lt;br /&gt;
** Cross-validation: Particularly useful for predictive performance.&lt;br /&gt;
&lt;br /&gt;
* Handling Overdispersion: For Poisson models: \( \text{Var}(Y) = \phi\mu \) where \(\phi &amp;gt; 1\) indicates overdispersion&lt;br /&gt;
Solutions:&lt;br /&gt;
** Use quasi-Poisson model&lt;br /&gt;
** Use negative binomial distribution&lt;br /&gt;
** Include random effects.&lt;br /&gt;
&lt;br /&gt;
* Zero-Inflation: For count data with excess zeros, consider:&lt;br /&gt;
** Zero-inflated Poisson (ZIP) model&lt;br /&gt;
** Zero-inflated negative binomial (ZINB) model&lt;br /&gt;
** Hurdle models.&lt;br /&gt;
&lt;br /&gt;
===Advanced Topics===&lt;br /&gt;
&lt;br /&gt;
==== Mixed Effects GLM (GLMM)====&lt;br /&gt;
&lt;br /&gt;
Extension incorporating random effects:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
g(E[Y|b]) = \mathbf{X}\beta + \mathbf{Z}b&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \( b \sim N(0, \mathbf{G}) \).&lt;br /&gt;
&lt;br /&gt;
Implementation in R with &amp;lt;code&amp;gt;lme4::glmer()&amp;lt;/code&amp;gt;:&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(lme4)&lt;br /&gt;
&lt;br /&gt;
# generate random data&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
id &amp;lt;- seq(n)&lt;br /&gt;
day &amp;lt;- 1:20&lt;br /&gt;
mydata &amp;lt;- expand.grid(id = id, day = day)&lt;br /&gt;
set.seed(1)&lt;br /&gt;
trt &amp;lt;- sample(c(&amp;quot;control&amp;quot;, &amp;quot;treat&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
sex &amp;lt;- sample(c(&amp;quot;female&amp;quot;, &amp;quot;male&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
mydata$trt &amp;lt;- trt[mydata$id]&lt;br /&gt;
mydata$sex &amp;lt;- sex[mydata$id]&lt;br /&gt;
mydata &amp;lt;- mydata[order(mydata$id, mydata$day),]&lt;br /&gt;
rownames(mydata) &amp;lt;- NULL&lt;br /&gt;
head(mydata, n = 10)&lt;br /&gt;
&lt;br /&gt;
mydata$trtsex &amp;lt;- interaction(mydata$trt, mydata$sex)&lt;br /&gt;
probs &amp;lt;- c(0.40, 0.85, 0.30, 0.50)&lt;br /&gt;
names(probs) &amp;lt;- levels(mydata$trtsex)&lt;br /&gt;
mydata$p &amp;lt;- probs[mydata$trtsex]&lt;br /&gt;
&lt;br /&gt;
set.seed(3)&lt;br /&gt;
r_probs &amp;lt;- rnorm(n = n, mean = 0, sd = 0.03)&lt;br /&gt;
mydata$random_p &amp;lt;- r_probs[mydata$id]&lt;br /&gt;
mydata$p &amp;lt;- mydata$p + mydata$random_p&lt;br /&gt;
&lt;br /&gt;
# use the probabilities to generate zeroes and ones with the binom() function.&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n = nrow(mydata), size = 1, prob = mydata$p)&lt;br /&gt;
&lt;br /&gt;
# using the sim data, inspect the first few records.&lt;br /&gt;
&lt;br /&gt;
head(mydata[c(&amp;quot;id&amp;quot;, &amp;quot;day&amp;quot;, &amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;, &amp;quot;p&amp;quot;, &amp;quot;y&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
# Example with binary outcome and random intercept&lt;br /&gt;
m &amp;lt;- glmer(y ~ trt * sex + (1|id), data = mydata, family = binomial)&lt;br /&gt;
summary(m, corr = FALSE)&lt;br /&gt;
&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Bayesian GLM====&lt;br /&gt;
&lt;br /&gt;
Using Markov Chain Monte Carlo (MCMC) methods.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
# Bayesian logistic regression&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(y ~ day +  trt + sex + p, # predictors,&lt;br /&gt;
                        family = binomial,&lt;br /&gt;
                        data = mydata,&lt;br /&gt;
                        prior = normal(0, 2.5),&lt;br /&gt;
                        prior_intercept = normal(0, 5))&lt;br /&gt;
&lt;br /&gt;
print(bayes_model, digits = 2)&lt;br /&gt;
&lt;br /&gt;
# Coefficient Plot (Forest Plot)&lt;br /&gt;
&lt;br /&gt;
plot(bayes_model, &amp;quot;areas&amp;quot;)  # density + intervals&lt;br /&gt;
# OR more customizable:&lt;br /&gt;
library(bayesplot)&lt;br /&gt;
mcmc_intervals(bayes_model, prob = 0.95) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Distributions of Coefficients&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Posterior Predictive Checks (PPC)&lt;br /&gt;
# Check if the model can reproduce data like the observed&lt;br /&gt;
pp_check(bayes_model, plotfun = &amp;quot;hist&amp;quot;, nreps = 100) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Predictive Check (Histogram of y_rep vs y)&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Another Bayesian Experiment.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
set.seed(123)&lt;br /&gt;
&lt;br /&gt;
# Simulate data — explicitly make trt and sex into factors&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
mydata &amp;lt;- data.frame(&lt;br /&gt;
  day = sample(1:30, n, replace = TRUE),&lt;br /&gt;
  trt = factor(sample(c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;)),&lt;br /&gt;
  sex = factor(sample(c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;)),&lt;br /&gt;
  p   = rnorm(n, mean = 50, sd = 10)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# True log-odds&lt;br /&gt;
logit_p &amp;lt;- -1 - 0.02 * mydata$day + &lt;br /&gt;
           0.6 * (as.numeric(mydata$trt) - 1) +   # A=0, B=1&lt;br /&gt;
           0.3 * (as.numeric(mydata$sex) - 1) +    # F=0, M=1&lt;br /&gt;
           0.04 * mydata$p&lt;br /&gt;
&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n, size = 1, prob = plogis(logit_p))&lt;br /&gt;
&lt;br /&gt;
# Fit model&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(&lt;br /&gt;
  y ~ day + trt + sex + p,&lt;br /&gt;
  family = binomial,&lt;br /&gt;
  data = mydata,&lt;br /&gt;
  prior = normal(0, 2.5),&lt;br /&gt;
  prior_intercept = normal(0, 5),&lt;br /&gt;
  seed = 42,&lt;br /&gt;
  refresh = 0&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Create New Data&lt;br /&gt;
# Confirm variables are factors&lt;br /&gt;
str(mydata[c(&amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
# Create prediction grid&lt;br /&gt;
newdata &amp;lt;- expand.grid(&lt;br /&gt;
  day = seq(min(mydata$day), max(mydata$day), length.out = 30),&lt;br /&gt;
  trt = levels(mydata$trt)[1],      # e.g., &amp;quot;A&amp;quot;&lt;br /&gt;
  sex = levels(mydata$sex)[1],      # e.g., &amp;quot;F&amp;quot;&lt;br /&gt;
  p   = median(mydata$p)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Verify it worked&lt;br /&gt;
stopifnot(nrow(newdata) &amp;gt; 0)&lt;br /&gt;
str(newdata)&lt;br /&gt;
&lt;br /&gt;
# Generate Predictions from the Bayesian Posterior Probability (after fitting the Bayesian Model):&lt;br /&gt;
# Posterior predicted probabilities&lt;br /&gt;
post_pred &amp;lt;- posterior_linpred(bayes_model, newdata = newdata, transform = TRUE)&lt;br /&gt;
&lt;br /&gt;
# Summarize&lt;br /&gt;
pred_summary &amp;lt;- data.frame(&lt;br /&gt;
  day = newdata$day,&lt;br /&gt;
  prob = apply(post_pred, 2, median),&lt;br /&gt;
  lower = apply(post_pred, 2, quantile, 0.025),&lt;br /&gt;
  upper = apply(post_pred, 2, quantile, 0.975)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Plot&lt;br /&gt;
library(ggplot2)&lt;br /&gt;
ggplot(pred_summary, aes(x = day, y = prob)) +&lt;br /&gt;
  geom_line(color = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  geom_ribbon(aes(ymin = lower, ymax = upper), alpha = 0.2, fill = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  labs(&lt;br /&gt;
    title = &amp;quot;Predicted Probability of Outcome by Day&amp;quot;,&lt;br /&gt;
    subtitle = &amp;quot;Holding treatment = A, sex = F, and p = median&amp;quot;,&lt;br /&gt;
    y = &amp;quot;P(y = 1)&amp;quot;,&lt;br /&gt;
    x = &amp;quot;Day&amp;quot;&lt;br /&gt;
  ) +&lt;br /&gt;
  theme_minimal()&lt;br /&gt;
&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Problems===&lt;br /&gt;
&lt;br /&gt;
1. Conceptual Exercises:&lt;br /&gt;
&lt;br /&gt;
   a) Derive the score equations for a Poisson GLM with log link&lt;br /&gt;
   b) Show that the binomial distribution with logit link is the canonical link&lt;br /&gt;
   c) Prove that the deviance for a normal GLM equals the residual sum of squares&lt;br /&gt;
&lt;br /&gt;
2. Applied Problems:&lt;br /&gt;
&lt;br /&gt;
   a) Analyze the [http://wiki.socr.umich.edu/index.php/SOCR_Data_Dinov_021808_ConsumerPriceIndex3Way Consumer Price Index] data using appropriate GLM&lt;br /&gt;
   b) Model the [http://wiki.socr.umich.edu/index.php/SOCR_Data_MonetaryBase1959_2009 Monetary Base] data considering temporal autocorrelation&lt;br /&gt;
   c) Using the &amp;lt;code&amp;gt;iris&amp;lt;/code&amp;gt; dataset, build a multinomial logistic regression to classify species&lt;br /&gt;
&lt;br /&gt;
3. Practice Simulation Study:&lt;br /&gt;
&lt;br /&gt;
   &amp;lt;pre&amp;gt;&lt;br /&gt;
   # Simulate data from a logistic regression model&lt;br /&gt;
   set.seed(123)&lt;br /&gt;
   n &amp;lt;- 1000&lt;br /&gt;
   x1 &amp;lt;- rnorm(n)&lt;br /&gt;
   x2 &amp;lt;- rnorm(n)&lt;br /&gt;
   beta &amp;lt;- c(0.5, 1, -0.5)&lt;br /&gt;
   linear_predictor &amp;lt;- beta[1] + beta[2]*x1 + beta[3]*x2&lt;br /&gt;
   probabilities &amp;lt;- plogis(linear_predictor)&lt;br /&gt;
   y &amp;lt;- rbinom(n, size = 1, prob = probabilities)&lt;br /&gt;
   &lt;br /&gt;
   # Fit model and evaluate performance&lt;br /&gt;
   sim_data &amp;lt;- data.frame(y = y, x1 = x1, x2 = x2)&lt;br /&gt;
   model_sim &amp;lt;- glm(y ~ x1 + x2, family = binomial, data = sim_data)&lt;br /&gt;
   &lt;br /&gt;
   # Calculate bias and MSE&lt;br /&gt;
   beta_hat &amp;lt;- coef(model_sim)&lt;br /&gt;
   bias &amp;lt;- beta_hat - beta&lt;br /&gt;
   mse &amp;lt;- mean((beta_hat - beta)^2)&lt;br /&gt;
   &lt;br /&gt;
   cat(&amp;quot;True parameters:&amp;quot;, beta, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Estimated parameters:&amp;quot;, beta_hat, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Bias:&amp;quot;, bias, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;MSE:&amp;quot;, mse, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   &amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
&lt;br /&gt;
1. McCullagh, P., &amp;amp; Nelder, J. A. (1989). *Generalized Linear Models* (2nd ed.). Chapman and Hall.&lt;br /&gt;
&lt;br /&gt;
2. Dobson, A. J., &amp;amp; Barnett, A. G. (2018). *An Introduction to Generalized Linear Models* (4th ed.). CRC Press.&lt;br /&gt;
&lt;br /&gt;
3. Agresti, A. (2015). *Foundations of Linear and Generalized Linear Models*. Wiley.&lt;br /&gt;
&lt;br /&gt;
4. Fahrmeir, L., Kneib, T., Lang, S., &amp;amp; Marx, B. (2013). *Regression: Models, Methods and Applications*. Springer.&lt;br /&gt;
&lt;br /&gt;
5. Wood, S. N. (2017). *Generalized Additive Models: An Introduction with R* (2nd ed.). Chapman and Hall/CRC.&lt;br /&gt;
&lt;br /&gt;
====Online Resources====&lt;br /&gt;
* [https://en.wikipedia.org/wiki/Generalized_linear_model GLM Wikipedia]&lt;br /&gt;
* [https://data.princeton.edu/wws509/notes/a2.pdf GLM Theory Notes]&lt;br /&gt;
* [https://www.jstatsoft.org/article/view/v015i12 GLM in R Tutorial]&lt;br /&gt;
* [https://cran.r-project.org/web/views/SocialSciences.html R Resources for Social Sciences]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php?title=SMHS_GLM}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18352</id>
		<title>SMHS GLM</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18352"/>
		<updated>2026-03-10T16:56:49Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Example 1: Logistic Regression for Contraceptive Use */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Generalized Linear Modeling (GLM) ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Generalized Linear Modeling (GLM) is a flexible generalization of ordinary linear regression that allows response variables to have error distribution models other than a normal distribution. GLM extends linear regression by allowing the linear model to be related to the response variable via a link function and enabling the variance of each measurement to be a function of its predicted value. This framework unifies statistical models including linear regression, logistic regression, and Poisson regression. Estimation methods include iteratively reweighted least squares for maximum likelihood estimation, Bayesian approaches, and least squares fitted to variance-stabilized responses.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
While linear regression models linear relationships between response and predictors, many real-world scenarios involve response variables that don't follow normal distributions. For example:&lt;br /&gt;
* Binary outcomes (yes/no decisions) with probabilities bounded between 0 and 1&lt;br /&gt;
* Count data (number of events) that follow Poisson distributions&lt;br /&gt;
* Survival times that follow exponential or Weibull distributions&lt;br /&gt;
&lt;br /&gt;
GLM provides a unified framework for these situations by allowing response variables from the exponential family of distributions.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====1) GLM Components====&lt;br /&gt;
A GLM consists of three components:&lt;br /&gt;
&lt;br /&gt;
1. Random Component: The response variable \(Y\) follows a distribution from the exponential family:&lt;br /&gt;
   &amp;lt;math&amp;gt;f_Y(y|\theta,\phi) = \exp\left\{\frac{y\theta - b(\theta)}{a(\phi)} + c(y,\phi)\right\}&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;\theta&amp;lt;/math&amp;gt; is the natural parameter and \(\phi\) is the dispersion parameter.&lt;br /&gt;
&lt;br /&gt;
2. Systematic Component: The linear predictor \(\eta\):&lt;br /&gt;
   &amp;lt;math&amp;gt;\eta = \mathbf{X}\beta = \beta_0 + \beta_1X_1 + \beta_2X_2 + \cdots + \beta_pX_p&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
3. Link Function: \(g(\cdot)\) that relates the mean \(\mu = E[Y]\) to the linear predictor:&lt;br /&gt;
   &amp;lt;math&amp;gt;g(\mu) = \eta \quad \text{or equivalently} \quad \mu = g^{-1}(\eta)&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The variance function relates the variance to the mean: &amp;lt;math&amp;gt;\text{Var}(Y) = a(\phi)V(\mu)&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;V(\mu)&amp;lt;/math&amp;gt; is the variance function specific to the distribution.&lt;br /&gt;
&lt;br /&gt;
====2) Exponential Family Distributions====&lt;br /&gt;
The exponential family includes many common distributions:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Distribution !! Support !! Natural Parameter \(\theta\) !! \(b(\theta)\) !! Canonical Link \(g(\mu)\) !! Variance Function \(V(\mu)\)&lt;br /&gt;
|-&lt;br /&gt;
| Normal || \(\mathbb{R}\) || \(\mu\) || \(\frac{\theta^2}{2}\) || Identity: \(\mu\) || 1&lt;br /&gt;
|-&lt;br /&gt;
| Binomial || \(\{0,1,\ldots,n\}\) || \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(n\log(1+e^\theta)\) || Logit: \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(\mu\left(1-\frac{\mu}{n}\right)\)&lt;br /&gt;
|-&lt;br /&gt;
| Poisson || \(\mathbb{N}_0\) || \(\log(\mu)\) || \(e^\theta\) || Log: \(\log(\mu)\) || \(\mu\)&lt;br /&gt;
|-&lt;br /&gt;
| Gamma || \(\mathbb{R}^+\) || \(-\frac{1}{\mu}\) || \(-\log(-\theta)\) || Inverse: \(\frac{1}{\mu}\) || \(\mu^2\)&lt;br /&gt;
|-&lt;br /&gt;
| Inverse Gaussian || \(\mathbb{R}^+\) || \(-\frac{1}{2\mu^2}\) || \(-\sqrt{-2\theta}\) || Inverse squared: \(\frac{1}{\mu^2}\) || \(\mu^3\)&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====3) Maximum Likelihood Estimation====&lt;br /&gt;
For a GLM with \(n\) independent observations, the log-likelihood is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\ell(\beta) = \sum_{i=1}^n \frac{y_i\theta_i - b(\theta_i)}{a(\phi)} + c(y_i,\phi)&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\theta_i = \theta(\mu_i)\) and \(\mu_i = g^{-1}(\mathbf{x}_i^\top\beta)\).&lt;br /&gt;
&lt;br /&gt;
The score equations are:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathbf{U}(\beta) = \frac{\partial\ell}{\partial\beta} = \mathbf{X}^\top\mathbf{W}(\mathbf{y} - \boldsymbol{\mu}) = 0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\mathbf{W} = \text{diag}\left\{\frac{1}{a(\phi)V(\mu_i)[g'(\mu_i)]^2}\right\}\).&lt;br /&gt;
&lt;br /&gt;
The Fisher information matrix is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathcal{I}(\beta) = E\left[-\frac{\partial^2\ell}{\partial\beta\partial\beta^\top}\right] = \mathbf{X}^\top\mathbf{W}\mathbf{X}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Parameters are estimated via Iteratively Reweighted Least Squares (IRLS):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\beta^{(t+1)} = \beta^{(t)} + (\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{z}^{(t)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(z_i^{(t)} = \eta_i^{(t)} + (y_i - \mu_i^{(t)})g'(\mu_i^{(t)})\).&lt;br /&gt;
&lt;br /&gt;
====4) Deviance and Goodness-of-Fit====&lt;br /&gt;
The deviance measures goodness-of-fit:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D = 2[\ell(\text{saturated model}) - \ell(\text{fitted model})]&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
For nested models \(M_0 \subset M_1\):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D_{M_0} - D_{M_1} \sim \chi^2_{df_{M_0} - df_{M_1}} \quad \text{under } H_0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The scaled deviance is \(D^* = D/\phi\), and Pearson's chi-square statistic is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\chi^2_P = \sum_{i=1}^n \frac{(y_i - \hat{\mu}_i)^2}{V(\hat{\mu}_i)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====5) Model Diagnostics====&lt;br /&gt;
Key diagnostic tools:&lt;br /&gt;
* Pearson residuals: \(r_i^P = \frac{y_i - \hat{\mu}_i}{\sqrt{V(\hat{\mu}_i)}}\)&lt;br /&gt;
* Deviance residuals: \(r_i^D = \text{sign}(y_i - \hat{\mu}_i)\sqrt{d_i}\)&lt;br /&gt;
* Leverage: \(h_{ii}\) from the hat matrix \(\mathbf{H} = \mathbf{W}^{1/2}\mathbf{X}(\mathbf{X}^\top\mathbf{W}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{1/2}\)&lt;br /&gt;
* Cook's distance: \(D_i = \frac{r_i^2 h_{ii}}{p(1-h_{ii})}\)&lt;br /&gt;
&lt;br /&gt;
===Applications===&lt;br /&gt;
&lt;br /&gt;
====Example 1: Logistic Regression for Contraceptive Use====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Load and prepare data&lt;br /&gt;
# https://grodri.github.io/datasets/cuse.dat&lt;br /&gt;
cuse &amp;lt;- read.table(&amp;quot;https://grodri.github.io/datasets/cuse.dat&amp;quot;, header=TRUE)&lt;br /&gt;
cat(&amp;quot;First few rows of dataset:\n&amp;quot;)&lt;br /&gt;
print(head(cuse))&lt;br /&gt;
&lt;br /&gt;
# Fit binomial GLM with logit link&lt;br /&gt;
model1 &amp;lt;- glm(cbind(using, notUsing) ~ age + education + wantsMore,&lt;br /&gt;
              family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Confidence Intervals ===\n&amp;quot;)&lt;br /&gt;
confint(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Odds Ratios with 95% CI ===\n&amp;quot;)&lt;br /&gt;
exp_coef &amp;lt;- exp(coef(model1))&lt;br /&gt;
exp_ci &amp;lt;- exp(confint(model1))&lt;br /&gt;
cbind(OR = exp_coef, exp_ci)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Comparison (Likelihood Ratio Test) ===\n&amp;quot;)&lt;br /&gt;
# Reduced model without education&lt;br /&gt;
model_reduced &amp;lt;- glm(cbind(using, notUsing) ~ age + wantsMore,&lt;br /&gt;
                     family = binomial, data = cuse)&lt;br /&gt;
anova(model_reduced, model1, test = &amp;quot;Chisq&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Diagnostic plots&lt;br /&gt;
par(mfrow = c(2, 2))&lt;br /&gt;
plot(model1, which = 1:4)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
'''Result Interpretation''':&lt;br /&gt;
&lt;br /&gt;
Let's explicate ''how to interpret the parameter estimates in this GLM model''.&lt;br /&gt;
Specifically, as this is a bivariate outcome, the estimates are not correlations&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
glm(formula = cbind(using, notUsing) ~ age + education + wantsMore, &lt;br /&gt;
    family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
Coefficients:&lt;br /&gt;
             Estimate Std. Error z value Pr(&amp;gt;|z|)    &lt;br /&gt;
(Intercept)   -0.8082     0.1590  -5.083 3.71e-07 ***&lt;br /&gt;
age25-29       0.3894     0.1759   2.214  0.02681 *  &lt;br /&gt;
age30-39       0.9086     0.1646   5.519 3.40e-08 ***&lt;br /&gt;
age40-49       1.1892     0.2144   5.546 2.92e-08 ***&lt;br /&gt;
educationlow  -0.3250     0.1240  -2.620  0.00879 ** &lt;br /&gt;
wantsMoreyes  -0.8330     0.1175  -7.091 1.33e-12 ***&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
In a ''Generalized Linear Model (GLM)'' with a ''binomial family'' and a ''logit link'' (the default for ''R''), the coefficients represent '''log-odds ratios'''.&lt;br /&gt;
Since our outcome is structured as &lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
   cbind(using, notUsing),&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
we are modeling the probability of &amp;quot;using&amp;quot; (success) versus &amp;quot;not using&amp;quot; (failure).&lt;br /&gt;
&lt;br /&gt;
That is, the model interpretation is in terms of '''Log-Odds'''.&lt;br /&gt;
In this model, the relationship between the ''covariate predictors'' and the ''outcome'' is defined by the ''logit'' function&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\text{logit}(p) = \ln\left(\frac{p}{1-p}\right) = \beta_0 + \beta_1X_1 + \dots + \beta_kX_k.&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The estimate ''Sign'' matters:&lt;br /&gt;
&lt;br /&gt;
: ''Positive Estimate'': As the predictor increases (or if the category is present), the probability of &amp;quot;using&amp;quot; increases.&lt;br /&gt;
: ''Negative Estimate'': As the predictor increases, the probability of &amp;quot;using&amp;quot; decreases.&lt;br /&gt;
&lt;br /&gt;
The estimate ''Magnitude'' is interpreted via the &amp;quot;Exponentiation Trick&amp;quot;.&lt;br /&gt;
Because humans don't think naturally in &amp;quot;log-odds&amp;quot; terms, we usually ''exponentiate'' the coefficients (&amp;lt;math&amp;gt;e^\beta&amp;lt;/math&amp;gt;) to get the ''Odds Ratios (OR)''.&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Variable&lt;br /&gt;
! Estimate (β)&lt;br /&gt;
! Odds Ratio (e&amp;lt;sup&amp;gt;β&amp;lt;/sup&amp;gt;)&lt;br /&gt;
! Interpretation&lt;br /&gt;
|-&lt;br /&gt;
| '''age40-49'''&lt;br /&gt;
| 1.1892&lt;br /&gt;
| ≈ 3.28&lt;br /&gt;
| Women aged 40-49 have '''3.28 times the odds''' of using contraception compared to the reference age group (likely &amp;lt;25).&lt;br /&gt;
|-&lt;br /&gt;
| '''wantsMoreyes'''&lt;br /&gt;
| -0.8330&lt;br /&gt;
| ≈ 0.43&lt;br /&gt;
| Women who want more children have '''57% lower odds''' (1 - 0.43) of using contraception than those who don't.&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
Specific Breakdown of the Results:&lt;br /&gt;
&lt;br /&gt;
* ''Age'' (Categorical): R has treated age as a ''factor''. The baseline (reference) group is the youngest group (''under 25'').&lt;br /&gt;
:: As age increases (&amp;lt;math&amp;gt;25-29 \ \to 30-39\  \to 40-49&amp;lt;/math&amp;gt;), the coefficients become increasingly positive (&amp;lt;math&amp;gt;0.38 \to 0.90 \to 1.18&amp;lt;/math&amp;gt;).&lt;br /&gt;
:: Meaning: Older women in this dataset are significantly more likely to use contraception than the youngest group.&lt;br /&gt;
&lt;br /&gt;
* Education:&lt;br /&gt;
:: Estimate: -0.3250&lt;br /&gt;
:: Meaning: Being in the ''low education'' group is associated with a decrease in the log-odds of using contraception compared to the ''high'' group. Specifically, their odds are about 28\% lower (&amp;lt;math&amp;gt;e^{-0.325} \approx 0.72&amp;lt;/math&amp;gt;).&lt;br /&gt;
&lt;br /&gt;
* Wants More Children:&lt;br /&gt;
:: Estimate: -0.8330&lt;br /&gt;
:: Meaning: This is a strong negative predictor. If a woman wants more children, the probability of her using contraception drops significantly.&lt;br /&gt;
&lt;br /&gt;
We can have a quick &amp;quot;reality check&amp;quot; on the measuring units. Unlike a standard correlation (which is bounded between -1 and 1), these estimates can be any real number.&lt;br /&gt;
&lt;br /&gt;
:: An estimate of 0 means the variable has no effect on the odds (Odds Ratio = 1).&lt;br /&gt;
:: The z-value tells you how many standard errors the estimate is from zero. Since all your p-values are small (&amp;lt;math&amp;gt;&amp;lt; 0.05&amp;lt;/math&amp;gt;), all these predictors are statistically significant.&lt;br /&gt;
&lt;br /&gt;
====Example 2: Poisson Regression for Count Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with built-in R dataset: AIDS cases in Belgium&lt;br /&gt;
# Load necessary libraries&lt;br /&gt;
if (!require(&amp;quot;MASS&amp;quot;)) install.packages(&amp;quot;MASS&amp;quot;)&lt;br /&gt;
library(MASS)&lt;br /&gt;
&lt;br /&gt;
# Load AIDS dataset&lt;br /&gt;
data(Aids2)&lt;br /&gt;
cat(&amp;quot;AIDS dataset structure:\n&amp;quot;)&lt;br /&gt;
str(Aids2)&lt;br /&gt;
&lt;br /&gt;
# Prepare data: count of AIDS cases by year and state&lt;br /&gt;
aids_counts &amp;lt;- aggregate(cbind(count = 1:nrow(Aids2)) ~ age + state,&lt;br /&gt;
                         data = Aids2, FUN = length)&lt;br /&gt;
&lt;br /&gt;
# Fit Poisson regression&lt;br /&gt;
model2 &amp;lt;- glm(count ~ age + state, family = poisson, data = aids_counts)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Poisson Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model2)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Check for Overdispersion ===\n&amp;quot;)&lt;br /&gt;
# Pearson chi-square statistic&lt;br /&gt;
pearson_chi2 &amp;lt;- sum(residuals(model2, type = &amp;quot;pearson&amp;quot;)^2)&lt;br /&gt;
df_resid &amp;lt;- df.residual(model2)&lt;br /&gt;
dispersion &amp;lt;- pearson_chi2 / df_resid&lt;br /&gt;
cat(&amp;quot;Pearson χ²:&amp;quot;, pearson_chi2, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;Dispersion parameter:&amp;quot;, dispersion, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;p-value for H0: φ=1:&amp;quot;, pchisq(pearson_chi2, df_resid, lower.tail = FALSE), &amp;quot;\n&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# If overdispersed, fit quasipoisson model&lt;br /&gt;
if (dispersion &amp;gt; 1.5) {&lt;br /&gt;
  cat(&amp;quot;\n=== Fitting Quasi-Poisson Model (accounting for overdispersion) ===\n&amp;quot;)&lt;br /&gt;
  model2_qp &amp;lt;- glm(count ~ age + state, family = quasipoisson, data = aids_counts)&lt;br /&gt;
  summary(model2_qp)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Predict expected counts&lt;br /&gt;
cat(&amp;quot;\n=== Predictions for First 10 Observations ===\n&amp;quot;)&lt;br /&gt;
predictions &amp;lt;- predict(model2, type = &amp;quot;response&amp;quot;, se.fit = TRUE)&lt;br /&gt;
pred_df &amp;lt;- data.frame(&lt;br /&gt;
  Observed = aids_counts$count[1:10],&lt;br /&gt;
  Predicted = predictions$fit[1:10],&lt;br /&gt;
  SE = predictions$se.fit[1:10],&lt;br /&gt;
  Lower = predictions$fit[1:10] - 1.96 * predictions$se.fit[1:10],&lt;br /&gt;
  Upper = predictions$fit[1:10] + 1.96 * predictions$se.fit[1:10]&lt;br /&gt;
)&lt;br /&gt;
print(pred_df)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Example 3: Gamma Regression for Positive Continuous Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with insurance claims data&lt;br /&gt;
if (!require(&amp;quot;insuranceData&amp;quot;)) install.packages(&amp;quot;insuranceData&amp;quot;)&lt;br /&gt;
library(insuranceData)&lt;br /&gt;
data(dataCar)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;Car insurance claims dataset:\n&amp;quot;)&lt;br /&gt;
str(dataCar)&lt;br /&gt;
&lt;br /&gt;
# Filter for positive claim amounts&lt;br /&gt;
claims_positive &amp;lt;- subset(dataCar, claimcst0 &amp;gt; 0)&lt;br /&gt;
&lt;br /&gt;
# Fit Gamma GLM with log link (common for monetary amounts)&lt;br /&gt;
model3 &amp;lt;- glm(claimcst0 ~ agecat + area + veh_age,&lt;br /&gt;
              family = Gamma(link = &amp;quot;log&amp;quot;),&lt;br /&gt;
              data = claims_positive)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Gamma Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model3)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Checking Gamma Model Assumptions ===\n&amp;quot;)&lt;br /&gt;
# Check residuals&lt;br /&gt;
res_gamma &amp;lt;- residuals(model3, type = &amp;quot;deviance&amp;quot;)&lt;br /&gt;
par(mfrow = c(1, 2))&lt;br /&gt;
hist(res_gamma, main = &amp;quot;Deviance Residuals&amp;quot;, xlab = &amp;quot;Residuals&amp;quot;)&lt;br /&gt;
qqnorm(res_gamma, main = &amp;quot;Q-Q Plot of Residuals&amp;quot;)&lt;br /&gt;
qqline(res_gamma)&lt;br /&gt;
&lt;br /&gt;
# Scale-location plot&lt;br /&gt;
fitted_values &amp;lt;- fitted(model3)&lt;br /&gt;
plot(fitted_values, sqrt(abs(res_gamma)),&lt;br /&gt;
     xlab = &amp;quot;Fitted Values&amp;quot;, ylab = &amp;quot;√|Deviance Residuals|&amp;quot;,&lt;br /&gt;
     main = &amp;quot;Scale-Location Plot&amp;quot;)&lt;br /&gt;
lines(lowess(fitted_values, sqrt(abs(res_gamma))), col = &amp;quot;red&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Software Implementation in R===&lt;br /&gt;
&lt;br /&gt;
Note that `family` is a function (a &amp;quot;closure&amp;quot;), not a list or an object with a \(\$family\) component. In thecode, we're passing&lt;br /&gt;
the family function itself (e.g., gaussian) to the \(run\_glm\_analysis()\) function, and later accessing \(family\$family\), &lt;br /&gt;
which doesn’t exist—because gaussian is a function, not a fitted model object.&lt;br /&gt;
In R, gaussian, binomial, poisson, etc., are functions that return family objects when called.&lt;br /&gt;
We're passing the function, not the result of calling it.&lt;br /&gt;
However, `glm()` internally calls `family()`, i.e., `gaussian()`, to get the actual family list object, which does have a \(\$family\) element.&lt;br /&gt;
Instead of checking \(family\$family\), we extract the family name from the ''fitted model'', not from the input argument.&lt;br /&gt;
Specifically, after fitting the model, we use \(model\$family\$family\), which is a character string like &amp;quot;gaussian&amp;quot;, &amp;quot;binomial&amp;quot;, etc.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Comprehensive GLM analysis function&lt;br /&gt;
run_glm_analysis &amp;lt;- function(formula, data, family, link = NULL) {&lt;br /&gt;
  # Fit the model&lt;br /&gt;
  if (!is.null(link)) {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family(link = link))&lt;br /&gt;
  } else {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family)&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  # Model summary&lt;br /&gt;
  cat(&amp;quot;=== MODEL SUMMARY ===\n&amp;quot;)&lt;br /&gt;
  print(summary(model))&lt;br /&gt;
  &lt;br /&gt;
  # Confidence intervals&lt;br /&gt;
  cat(&amp;quot;\n=== 95% CONFIDENCE INTERVALS ===\n&amp;quot;)&lt;br /&gt;
  print(confint(model))&lt;br /&gt;
  &lt;br /&gt;
  # Goodness-of-fit tests&lt;br /&gt;
  cat(&amp;quot;\n=== GOODNESS-OF-FIT ===\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Null Deviance:&amp;quot;, model$null.deviance, &amp;quot;on&amp;quot;, model$df.null, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Residual Deviance:&amp;quot;, model$deviance, &amp;quot;on&amp;quot;, model$df.residual, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;AIC:&amp;quot;, AIC(model), &amp;quot;\n&amp;quot;)&lt;br /&gt;
  &lt;br /&gt;
  # Check for overdispersion (for Poisson and binomial families)&lt;br /&gt;
  family_name &amp;lt;- model$family$family  # ✅ Extract from fitted model&lt;br /&gt;
  if (family_name %in% c(&amp;quot;poisson&amp;quot;, &amp;quot;binomial&amp;quot;, &amp;quot;quasipoisson&amp;quot;, &amp;quot;quasibinomial&amp;quot;)) {&lt;br /&gt;
    cat(&amp;quot;\n=== OVERDISPERSION CHECK ===\n&amp;quot;)&lt;br /&gt;
    dispersion &amp;lt;- model$deviance / model$df.residual&lt;br /&gt;
    cat(&amp;quot;Dispersion parameter:&amp;quot;, round(dispersion, 4), &amp;quot;\n&amp;quot;)&lt;br /&gt;
    if (abs(dispersion - 1) &amp;gt; 0.1) {&lt;br /&gt;
      direction &amp;lt;- ifelse(dispersion &amp;gt; 1, &amp;quot;over&amp;quot;, &amp;quot;under&amp;quot;)&lt;br /&gt;
      cat(&amp;quot;Note: Significant&amp;quot;, direction, &amp;quot;dispersion detected\n&amp;quot;)&lt;br /&gt;
    }&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  return(model)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Example usage with mtcars dataset&lt;br /&gt;
cat(&amp;quot;\n\n=== EXAMPLE: GAUSSIAN GLM (equivalent to linear regression) ===\n&amp;quot;)&lt;br /&gt;
data(mtcars)&lt;br /&gt;
model_gaussian &amp;lt;- run_glm_analysis(&lt;br /&gt;
  formula = mpg ~ wt + hp + cyl,&lt;br /&gt;
  data = mtcars,&lt;br /&gt;
  family = gaussian,&lt;br /&gt;
  link = &amp;quot;identity&amp;quot;&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Compare with lm() for verification&lt;br /&gt;
cat(&amp;quot;\n=== COMPARISON WITH lm() ===\n&amp;quot;)&lt;br /&gt;
model_lm &amp;lt;- lm(mpg ~ wt + hp + cyl, data = mtcars)&lt;br /&gt;
print(summary(model_lm))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Common GLM Families and Links in R===&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Family !! Default Link !! Alternative Links !! Typical Use&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || Continuous, symmetric data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;binomial()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;logit&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;probit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cauchit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cloglog&amp;lt;/code&amp;gt; || Binary/count proportions&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;poisson()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;sqrt&amp;lt;/code&amp;gt; || Count data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;Gamma()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, right-skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;inverse.gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;1/μ²&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, highly skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;quasi()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || User-defined || Overdispersed data&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Practical Considerations===&lt;br /&gt;
&lt;br /&gt;
* Model Selection:&lt;br /&gt;
** AIC (Akaike Information Criterion): \( \text{AIC} = -2\ell + 2p \)&lt;br /&gt;
** BIC (Bayesian Information Criterion): \( \text{BIC} = -2\ell + p\log(n) \)&lt;br /&gt;
** Cross-validation: Particularly useful for predictive performance.&lt;br /&gt;
&lt;br /&gt;
* Handling Overdispersion: For Poisson models: \( \text{Var}(Y) = \phi\mu \) where \(\phi &amp;gt; 1\) indicates overdispersion&lt;br /&gt;
Solutions:&lt;br /&gt;
** Use quasi-Poisson model&lt;br /&gt;
** Use negative binomial distribution&lt;br /&gt;
** Include random effects.&lt;br /&gt;
&lt;br /&gt;
* Zero-Inflation: For count data with excess zeros, consider:&lt;br /&gt;
** Zero-inflated Poisson (ZIP) model&lt;br /&gt;
** Zero-inflated negative binomial (ZINB) model&lt;br /&gt;
** Hurdle models.&lt;br /&gt;
&lt;br /&gt;
===Advanced Topics===&lt;br /&gt;
&lt;br /&gt;
==== Mixed Effects GLM (GLMM)====&lt;br /&gt;
&lt;br /&gt;
Extension incorporating random effects:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
g(E[Y|b]) = \mathbf{X}\beta + \mathbf{Z}b&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \( b \sim N(0, \mathbf{G}) \).&lt;br /&gt;
&lt;br /&gt;
Implementation in R with &amp;lt;code&amp;gt;lme4::glmer()&amp;lt;/code&amp;gt;:&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(lme4)&lt;br /&gt;
&lt;br /&gt;
# generate random data&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
id &amp;lt;- seq(n)&lt;br /&gt;
day &amp;lt;- 1:20&lt;br /&gt;
mydata &amp;lt;- expand.grid(id = id, day = day)&lt;br /&gt;
set.seed(1)&lt;br /&gt;
trt &amp;lt;- sample(c(&amp;quot;control&amp;quot;, &amp;quot;treat&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
sex &amp;lt;- sample(c(&amp;quot;female&amp;quot;, &amp;quot;male&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
mydata$trt &amp;lt;- trt[mydata$id]&lt;br /&gt;
mydata$sex &amp;lt;- sex[mydata$id]&lt;br /&gt;
mydata &amp;lt;- mydata[order(mydata$id, mydata$day),]&lt;br /&gt;
rownames(mydata) &amp;lt;- NULL&lt;br /&gt;
head(mydata, n = 10)&lt;br /&gt;
&lt;br /&gt;
mydata$trtsex &amp;lt;- interaction(mydata$trt, mydata$sex)&lt;br /&gt;
probs &amp;lt;- c(0.40, 0.85, 0.30, 0.50)&lt;br /&gt;
names(probs) &amp;lt;- levels(mydata$trtsex)&lt;br /&gt;
mydata$p &amp;lt;- probs[mydata$trtsex]&lt;br /&gt;
&lt;br /&gt;
set.seed(3)&lt;br /&gt;
r_probs &amp;lt;- rnorm(n = n, mean = 0, sd = 0.03)&lt;br /&gt;
mydata$random_p &amp;lt;- r_probs[mydata$id]&lt;br /&gt;
mydata$p &amp;lt;- mydata$p + mydata$random_p&lt;br /&gt;
&lt;br /&gt;
# use the probabilities to generate zeroes and ones with the binom() function.&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n = nrow(mydata), size = 1, prob = mydata$p)&lt;br /&gt;
&lt;br /&gt;
# using the sim data, inspect the first few records.&lt;br /&gt;
&lt;br /&gt;
head(mydata[c(&amp;quot;id&amp;quot;, &amp;quot;day&amp;quot;, &amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;, &amp;quot;p&amp;quot;, &amp;quot;y&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
# Example with binary outcome and random intercept&lt;br /&gt;
m &amp;lt;- glmer(y ~ trt * sex + (1|id), data = mydata, family = binomial)&lt;br /&gt;
summary(m, corr = FALSE)&lt;br /&gt;
&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Bayesian GLM====&lt;br /&gt;
&lt;br /&gt;
Using Markov Chain Monte Carlo (MCMC) methods.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
# Bayesian logistic regression&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(y ~ day +  trt + sex + p, # predictors,&lt;br /&gt;
                        family = binomial,&lt;br /&gt;
                        data = mydata,&lt;br /&gt;
                        prior = normal(0, 2.5),&lt;br /&gt;
                        prior_intercept = normal(0, 5))&lt;br /&gt;
&lt;br /&gt;
print(bayes_model, digits = 2)&lt;br /&gt;
&lt;br /&gt;
# Coefficient Plot (Forest Plot)&lt;br /&gt;
&lt;br /&gt;
plot(bayes_model, &amp;quot;areas&amp;quot;)  # density + intervals&lt;br /&gt;
# OR more customizable:&lt;br /&gt;
library(bayesplot)&lt;br /&gt;
mcmc_intervals(bayes_model, prob = 0.95) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Distributions of Coefficients&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Posterior Predictive Checks (PPC)&lt;br /&gt;
# Check if the model can reproduce data like the observed&lt;br /&gt;
pp_check(bayes_model, plotfun = &amp;quot;hist&amp;quot;, nreps = 100) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Predictive Check (Histogram of y_rep vs y)&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Another Bayesian Experiment.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
set.seed(123)&lt;br /&gt;
&lt;br /&gt;
# Simulate data — explicitly make trt and sex into factors&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
mydata &amp;lt;- data.frame(&lt;br /&gt;
  day = sample(1:30, n, replace = TRUE),&lt;br /&gt;
  trt = factor(sample(c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;)),&lt;br /&gt;
  sex = factor(sample(c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;)),&lt;br /&gt;
  p   = rnorm(n, mean = 50, sd = 10)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# True log-odds&lt;br /&gt;
logit_p &amp;lt;- -1 - 0.02 * mydata$day + &lt;br /&gt;
           0.6 * (as.numeric(mydata$trt) - 1) +   # A=0, B=1&lt;br /&gt;
           0.3 * (as.numeric(mydata$sex) - 1) +    # F=0, M=1&lt;br /&gt;
           0.04 * mydata$p&lt;br /&gt;
&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n, size = 1, prob = plogis(logit_p))&lt;br /&gt;
&lt;br /&gt;
# Fit model&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(&lt;br /&gt;
  y ~ day + trt + sex + p,&lt;br /&gt;
  family = binomial,&lt;br /&gt;
  data = mydata,&lt;br /&gt;
  prior = normal(0, 2.5),&lt;br /&gt;
  prior_intercept = normal(0, 5),&lt;br /&gt;
  seed = 42,&lt;br /&gt;
  refresh = 0&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Create New Data&lt;br /&gt;
# Confirm variables are factors&lt;br /&gt;
str(mydata[c(&amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
# Create prediction grid&lt;br /&gt;
newdata &amp;lt;- expand.grid(&lt;br /&gt;
  day = seq(min(mydata$day), max(mydata$day), length.out = 30),&lt;br /&gt;
  trt = levels(mydata$trt)[1],      # e.g., &amp;quot;A&amp;quot;&lt;br /&gt;
  sex = levels(mydata$sex)[1],      # e.g., &amp;quot;F&amp;quot;&lt;br /&gt;
  p   = median(mydata$p)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Verify it worked&lt;br /&gt;
stopifnot(nrow(newdata) &amp;gt; 0)&lt;br /&gt;
str(newdata)&lt;br /&gt;
&lt;br /&gt;
# Generate Predictions from the Bayesian Posterior Probability (after fitting the Bayesian Model):&lt;br /&gt;
# Posterior predicted probabilities&lt;br /&gt;
post_pred &amp;lt;- posterior_linpred(bayes_model, newdata = newdata, transform = TRUE)&lt;br /&gt;
&lt;br /&gt;
# Summarize&lt;br /&gt;
pred_summary &amp;lt;- data.frame(&lt;br /&gt;
  day = newdata$day,&lt;br /&gt;
  prob = apply(post_pred, 2, median),&lt;br /&gt;
  lower = apply(post_pred, 2, quantile, 0.025),&lt;br /&gt;
  upper = apply(post_pred, 2, quantile, 0.975)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Plot&lt;br /&gt;
library(ggplot2)&lt;br /&gt;
ggplot(pred_summary, aes(x = day, y = prob)) +&lt;br /&gt;
  geom_line(color = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  geom_ribbon(aes(ymin = lower, ymax = upper), alpha = 0.2, fill = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  labs(&lt;br /&gt;
    title = &amp;quot;Predicted Probability of Outcome by Day&amp;quot;,&lt;br /&gt;
    subtitle = &amp;quot;Holding treatment = A, sex = F, and p = median&amp;quot;,&lt;br /&gt;
    y = &amp;quot;P(y = 1)&amp;quot;,&lt;br /&gt;
    x = &amp;quot;Day&amp;quot;&lt;br /&gt;
  ) +&lt;br /&gt;
  theme_minimal()&lt;br /&gt;
&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Problems===&lt;br /&gt;
&lt;br /&gt;
1. Conceptual Exercises:&lt;br /&gt;
&lt;br /&gt;
   a) Derive the score equations for a Poisson GLM with log link&lt;br /&gt;
   b) Show that the binomial distribution with logit link is the canonical link&lt;br /&gt;
   c) Prove that the deviance for a normal GLM equals the residual sum of squares&lt;br /&gt;
&lt;br /&gt;
2. Applied Problems:&lt;br /&gt;
&lt;br /&gt;
   a) Analyze the [http://wiki.socr.umich.edu/index.php/SOCR_Data_Dinov_021808_ConsumerPriceIndex3Way Consumer Price Index] data using appropriate GLM&lt;br /&gt;
   b) Model the [http://wiki.socr.umich.edu/index.php/SOCR_Data_MonetaryBase1959_2009 Monetary Base] data considering temporal autocorrelation&lt;br /&gt;
   c) Using the &amp;lt;code&amp;gt;iris&amp;lt;/code&amp;gt; dataset, build a multinomial logistic regression to classify species&lt;br /&gt;
&lt;br /&gt;
3. Practice Simulation Study:&lt;br /&gt;
&lt;br /&gt;
   &amp;lt;pre&amp;gt;&lt;br /&gt;
   # Simulate data from a logistic regression model&lt;br /&gt;
   set.seed(123)&lt;br /&gt;
   n &amp;lt;- 1000&lt;br /&gt;
   x1 &amp;lt;- rnorm(n)&lt;br /&gt;
   x2 &amp;lt;- rnorm(n)&lt;br /&gt;
   beta &amp;lt;- c(0.5, 1, -0.5)&lt;br /&gt;
   linear_predictor &amp;lt;- beta[1] + beta[2]*x1 + beta[3]*x2&lt;br /&gt;
   probabilities &amp;lt;- plogis(linear_predictor)&lt;br /&gt;
   y &amp;lt;- rbinom(n, size = 1, prob = probabilities)&lt;br /&gt;
   &lt;br /&gt;
   # Fit model and evaluate performance&lt;br /&gt;
   sim_data &amp;lt;- data.frame(y = y, x1 = x1, x2 = x2)&lt;br /&gt;
   model_sim &amp;lt;- glm(y ~ x1 + x2, family = binomial, data = sim_data)&lt;br /&gt;
   &lt;br /&gt;
   # Calculate bias and MSE&lt;br /&gt;
   beta_hat &amp;lt;- coef(model_sim)&lt;br /&gt;
   bias &amp;lt;- beta_hat - beta&lt;br /&gt;
   mse &amp;lt;- mean((beta_hat - beta)^2)&lt;br /&gt;
   &lt;br /&gt;
   cat(&amp;quot;True parameters:&amp;quot;, beta, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Estimated parameters:&amp;quot;, beta_hat, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Bias:&amp;quot;, bias, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;MSE:&amp;quot;, mse, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   &amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
&lt;br /&gt;
1. McCullagh, P., &amp;amp; Nelder, J. A. (1989). *Generalized Linear Models* (2nd ed.). Chapman and Hall.&lt;br /&gt;
&lt;br /&gt;
2. Dobson, A. J., &amp;amp; Barnett, A. G. (2018). *An Introduction to Generalized Linear Models* (4th ed.). CRC Press.&lt;br /&gt;
&lt;br /&gt;
3. Agresti, A. (2015). *Foundations of Linear and Generalized Linear Models*. Wiley.&lt;br /&gt;
&lt;br /&gt;
4. Fahrmeir, L., Kneib, T., Lang, S., &amp;amp; Marx, B. (2013). *Regression: Models, Methods and Applications*. Springer.&lt;br /&gt;
&lt;br /&gt;
5. Wood, S. N. (2017). *Generalized Additive Models: An Introduction with R* (2nd ed.). Chapman and Hall/CRC.&lt;br /&gt;
&lt;br /&gt;
====Online Resources====&lt;br /&gt;
* [https://en.wikipedia.org/wiki/Generalized_linear_model GLM Wikipedia]&lt;br /&gt;
* [https://data.princeton.edu/wws509/notes/a2.pdf GLM Theory Notes]&lt;br /&gt;
* [https://www.jstatsoft.org/article/view/v015i12 GLM in R Tutorial]&lt;br /&gt;
* [https://cran.r-project.org/web/views/SocialSciences.html R Resources for Social Sciences]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php?title=SMHS_GLM}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18351</id>
		<title>SMHS GLM</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SMHS_GLM&amp;diff=18351"/>
		<updated>2026-03-10T16:55:13Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Example 1: Logistic Regression for Contraceptive Use */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;==[[SMHS| Scientific Methods for Health Sciences]] - Generalized Linear Modeling (GLM) ==&lt;br /&gt;
&lt;br /&gt;
===Overview===&lt;br /&gt;
Generalized Linear Modeling (GLM) is a flexible generalization of ordinary linear regression that allows response variables to have error distribution models other than a normal distribution. GLM extends linear regression by allowing the linear model to be related to the response variable via a link function and enabling the variance of each measurement to be a function of its predicted value. This framework unifies statistical models including linear regression, logistic regression, and Poisson regression. Estimation methods include iteratively reweighted least squares for maximum likelihood estimation, Bayesian approaches, and least squares fitted to variance-stabilized responses.&lt;br /&gt;
&lt;br /&gt;
===Motivation===&lt;br /&gt;
While linear regression models linear relationships between response and predictors, many real-world scenarios involve response variables that don't follow normal distributions. For example:&lt;br /&gt;
* Binary outcomes (yes/no decisions) with probabilities bounded between 0 and 1&lt;br /&gt;
* Count data (number of events) that follow Poisson distributions&lt;br /&gt;
* Survival times that follow exponential or Weibull distributions&lt;br /&gt;
&lt;br /&gt;
GLM provides a unified framework for these situations by allowing response variables from the exponential family of distributions.&lt;br /&gt;
&lt;br /&gt;
===Theory===&lt;br /&gt;
&lt;br /&gt;
====1) GLM Components====&lt;br /&gt;
A GLM consists of three components:&lt;br /&gt;
&lt;br /&gt;
1. Random Component: The response variable \(Y\) follows a distribution from the exponential family:&lt;br /&gt;
   &amp;lt;math&amp;gt;f_Y(y|\theta,\phi) = \exp\left\{\frac{y\theta - b(\theta)}{a(\phi)} + c(y,\phi)\right\}&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;\theta&amp;lt;/math&amp;gt; is the natural parameter and \(\phi\) is the dispersion parameter.&lt;br /&gt;
&lt;br /&gt;
2. Systematic Component: The linear predictor \(\eta\):&lt;br /&gt;
   &amp;lt;math&amp;gt;\eta = \mathbf{X}\beta = \beta_0 + \beta_1X_1 + \beta_2X_2 + \cdots + \beta_pX_p&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
3. Link Function: \(g(\cdot)\) that relates the mean \(\mu = E[Y]\) to the linear predictor:&lt;br /&gt;
   &amp;lt;math&amp;gt;g(\mu) = \eta \quad \text{or equivalently} \quad \mu = g^{-1}(\eta)&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The variance function relates the variance to the mean: &amp;lt;math&amp;gt;\text{Var}(Y) = a(\phi)V(\mu)&amp;lt;/math&amp;gt;&lt;br /&gt;
where &amp;lt;math&amp;gt;V(\mu)&amp;lt;/math&amp;gt; is the variance function specific to the distribution.&lt;br /&gt;
&lt;br /&gt;
====2) Exponential Family Distributions====&lt;br /&gt;
The exponential family includes many common distributions:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Distribution !! Support !! Natural Parameter \(\theta\) !! \(b(\theta)\) !! Canonical Link \(g(\mu)\) !! Variance Function \(V(\mu)\)&lt;br /&gt;
|-&lt;br /&gt;
| Normal || \(\mathbb{R}\) || \(\mu\) || \(\frac{\theta^2}{2}\) || Identity: \(\mu\) || 1&lt;br /&gt;
|-&lt;br /&gt;
| Binomial || \(\{0,1,\ldots,n\}\) || \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(n\log(1+e^\theta)\) || Logit: \(\log\left(\frac{\mu}{n-\mu}\right)\) || \(\mu\left(1-\frac{\mu}{n}\right)\)&lt;br /&gt;
|-&lt;br /&gt;
| Poisson || \(\mathbb{N}_0\) || \(\log(\mu)\) || \(e^\theta\) || Log: \(\log(\mu)\) || \(\mu\)&lt;br /&gt;
|-&lt;br /&gt;
| Gamma || \(\mathbb{R}^+\) || \(-\frac{1}{\mu}\) || \(-\log(-\theta)\) || Inverse: \(\frac{1}{\mu}\) || \(\mu^2\)&lt;br /&gt;
|-&lt;br /&gt;
| Inverse Gaussian || \(\mathbb{R}^+\) || \(-\frac{1}{2\mu^2}\) || \(-\sqrt{-2\theta}\) || Inverse squared: \(\frac{1}{\mu^2}\) || \(\mu^3\)&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====3) Maximum Likelihood Estimation====&lt;br /&gt;
For a GLM with \(n\) independent observations, the log-likelihood is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\ell(\beta) = \sum_{i=1}^n \frac{y_i\theta_i - b(\theta_i)}{a(\phi)} + c(y_i,\phi)&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\theta_i = \theta(\mu_i)\) and \(\mu_i = g^{-1}(\mathbf{x}_i^\top\beta)\).&lt;br /&gt;
&lt;br /&gt;
The score equations are:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathbf{U}(\beta) = \frac{\partial\ell}{\partial\beta} = \mathbf{X}^\top\mathbf{W}(\mathbf{y} - \boldsymbol{\mu}) = 0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(\mathbf{W} = \text{diag}\left\{\frac{1}{a(\phi)V(\mu_i)[g'(\mu_i)]^2}\right\}\).&lt;br /&gt;
&lt;br /&gt;
The Fisher information matrix is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\mathcal{I}(\beta) = E\left[-\frac{\partial^2\ell}{\partial\beta\partial\beta^\top}\right] = \mathbf{X}^\top\mathbf{W}\mathbf{X}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Parameters are estimated via Iteratively Reweighted Least Squares (IRLS):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\beta^{(t+1)} = \beta^{(t)} + (\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{(t)}\mathbf{z}^{(t)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \(z_i^{(t)} = \eta_i^{(t)} + (y_i - \mu_i^{(t)})g'(\mu_i^{(t)})\).&lt;br /&gt;
&lt;br /&gt;
====4) Deviance and Goodness-of-Fit====&lt;br /&gt;
The deviance measures goodness-of-fit:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D = 2[\ell(\text{saturated model}) - \ell(\text{fitted model})]&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
For nested models \(M_0 \subset M_1\):&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
D_{M_0} - D_{M_1} \sim \chi^2_{df_{M_0} - df_{M_1}} \quad \text{under } H_0&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The scaled deviance is \(D^* = D/\phi\), and Pearson's chi-square statistic is:&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
\chi^2_P = \sum_{i=1}^n \frac{(y_i - \hat{\mu}_i)^2}{V(\hat{\mu}_i)}&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====5) Model Diagnostics====&lt;br /&gt;
Key diagnostic tools:&lt;br /&gt;
* Pearson residuals: \(r_i^P = \frac{y_i - \hat{\mu}_i}{\sqrt{V(\hat{\mu}_i)}}\)&lt;br /&gt;
* Deviance residuals: \(r_i^D = \text{sign}(y_i - \hat{\mu}_i)\sqrt{d_i}\)&lt;br /&gt;
* Leverage: \(h_{ii}\) from the hat matrix \(\mathbf{H} = \mathbf{W}^{1/2}\mathbf{X}(\mathbf{X}^\top\mathbf{W}\mathbf{X})^{-1}\mathbf{X}^\top\mathbf{W}^{1/2}\)&lt;br /&gt;
* Cook's distance: \(D_i = \frac{r_i^2 h_{ii}}{p(1-h_{ii})}\)&lt;br /&gt;
&lt;br /&gt;
===Applications===&lt;br /&gt;
&lt;br /&gt;
====Example 1: Logistic Regression for Contraceptive Use====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Load and prepare data&lt;br /&gt;
# https://grodri.github.io/datasets/cuse.dat&lt;br /&gt;
cuse &amp;lt;- read.table(&amp;quot;https://grodri.github.io/datasets/cuse.dat&amp;quot;, header=TRUE)&lt;br /&gt;
cat(&amp;quot;First few rows of dataset:\n&amp;quot;)&lt;br /&gt;
print(head(cuse))&lt;br /&gt;
&lt;br /&gt;
# Fit binomial GLM with logit link&lt;br /&gt;
model1 &amp;lt;- glm(cbind(using, notUsing) ~ age + education + wantsMore,&lt;br /&gt;
              family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Confidence Intervals ===\n&amp;quot;)&lt;br /&gt;
confint(model1)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Odds Ratios with 95% CI ===\n&amp;quot;)&lt;br /&gt;
exp_coef &amp;lt;- exp(coef(model1))&lt;br /&gt;
exp_ci &amp;lt;- exp(confint(model1))&lt;br /&gt;
cbind(OR = exp_coef, exp_ci)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Model Comparison (Likelihood Ratio Test) ===\n&amp;quot;)&lt;br /&gt;
# Reduced model without education&lt;br /&gt;
model_reduced &amp;lt;- glm(cbind(using, notUsing) ~ age + wantsMore,&lt;br /&gt;
                     family = binomial, data = cuse)&lt;br /&gt;
anova(model_reduced, model1, test = &amp;quot;Chisq&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Diagnostic plots&lt;br /&gt;
par(mfrow = c(2, 2))&lt;br /&gt;
plot(model1, which = 1:4)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
'''Result Interpretation''':&lt;br /&gt;
&lt;br /&gt;
Let's explicate ''how to interpret the parameter estimates in this GLM model''.&lt;br /&gt;
Specifically, as this is a bivariate outcome, the estimates are not correlations&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
glm(formula = cbind(using, notUsing) ~ age + education + wantsMore, &lt;br /&gt;
    family = binomial, data = cuse)&lt;br /&gt;
&lt;br /&gt;
Coefficients:&lt;br /&gt;
             Estimate Std. Error z value Pr(&amp;gt;|z|)    &lt;br /&gt;
(Intercept)   -0.8082     0.1590  -5.083 3.71e-07 ***&lt;br /&gt;
age25-29       0.3894     0.1759   2.214  0.02681 *  &lt;br /&gt;
age30-39       0.9086     0.1646   5.519 3.40e-08 ***&lt;br /&gt;
age40-49       1.1892     0.2144   5.546 2.92e-08 ***&lt;br /&gt;
educationlow  -0.3250     0.1240  -2.620  0.00879 ** &lt;br /&gt;
wantsMoreyes  -0.8330     0.1175  -7.091 1.33e-12 ***&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
In a ''Generalized Linear Model (GLM)'' with a ''binomial family'' and a ''logit link'' (the default for ''R''), the coefficients represent '''log-odds ratios'''.&lt;br /&gt;
Since our outcome is structured as &lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
   cbind(using, notUsing),&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
we are modeling the probability of &amp;quot;using&amp;quot; (success) versus &amp;quot;not using&amp;quot; (failure).&lt;br /&gt;
&lt;br /&gt;
That is, the model interpretation is in terms of '''Log-Odds'''.&lt;br /&gt;
In this model, the relationship between the ''covariate predictors'' and the ''outcome'' is defined by the ''logit'' function&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;\text{logit}(p) = \ln\left(\frac{p}{1-p}\right) = \beta_0 + \beta_1X_1 + \dots + \beta_kX_k.&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
&lt;br /&gt;
The estimate ''Sign'' matters:&lt;br /&gt;
&lt;br /&gt;
: ''Positive Estimate'': As the predictor increases (or if the category is present), the probability of &amp;quot;using&amp;quot; increases.&lt;br /&gt;
: ''Negative Estimate'': As the predictor increases, the probability of &amp;quot;using&amp;quot; decreases.&lt;br /&gt;
&lt;br /&gt;
The estimate ''Magnitude'' is interpreted via the &amp;quot;Exponentiation Trick&amp;quot;.&lt;br /&gt;
Because humans don't think naturally in &amp;quot;log-odds&amp;quot; terms, we usually ''exponentiate'' the coefficients (&amp;lt;math&amp;gt;e^\beta&amp;lt;/math&amp;gt;) to get the ''Odds Ratios (OR)''.&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
! Variable&lt;br /&gt;
! Estimate (β)&lt;br /&gt;
! Odds Ratio (e&amp;lt;sup&amp;gt;β&amp;lt;/sup&amp;gt;)&lt;br /&gt;
! Interpretation&lt;br /&gt;
|-&lt;br /&gt;
| '''age40-49'''&lt;br /&gt;
| 1.1892&lt;br /&gt;
| ≈ 3.28&lt;br /&gt;
| Women aged 40-49 have '''3.28 times the odds''' of using contraception compared to the reference age group (likely &amp;lt;25).&lt;br /&gt;
|-&lt;br /&gt;
| '''wantsMoreyes'''&lt;br /&gt;
| -0.8330&lt;br /&gt;
| ≈ 0.43&lt;br /&gt;
| Women who want more children have '''57% lower odds''' (1 - 0.43) of using contraception than those who don't.&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
Specific Breakdown of the Results:&lt;br /&gt;
&lt;br /&gt;
* ''Age'' (Categorical): R has treated age as a ''factor''. The baseline (reference) group is the youngest group (''under 25'').&lt;br /&gt;
:: As age increases (&amp;lt;math&amp;gt;25-29 \ \to 30-39\  \to 40-49&amp;lt;/math&amp;gt;), the coefficients become increasingly positive (&amp;lt;math&amp;gt;0.38 \to 0.90 \to 1.18&amp;lt;/math&amp;gt;).&lt;br /&gt;
:: Meaning: Older women in this dataset are significantly more likely to use contraception than the youngest group.&lt;br /&gt;
&lt;br /&gt;
* Education:&lt;br /&gt;
:: Estimate: -0.3250&lt;br /&gt;
:: Meaning: Being in the ''low education'' group is associated with a decrease in the log-odds of using contraception compared to the ''high'' group. Specifically, their odds are about 28\% lower (&amp;lt;math&amp;gt;e^{-0.325} \approx 0.72&amp;lt;\math&amp;gt;).&lt;br /&gt;
&lt;br /&gt;
* Wants More Children:&lt;br /&gt;
:: Estimate: -0.8330&lt;br /&gt;
:: Meaning: This is a strong negative predictor. If a woman wants more children, the probability of her using contraception drops significantly.&lt;br /&gt;
&lt;br /&gt;
We can have a quick &amp;quot;reality check&amp;quot; on the measuring units. Unlike a standard correlation (which is bounded between -1 and 1), these estimates can be any real number.&lt;br /&gt;
&lt;br /&gt;
:: An estimate of 0 means the variable has no effect on the odds (Odds Ratio = 1).&lt;br /&gt;
:: The z-value tells you how many standard errors the estimate is from zero. Since all your p-values are small (&amp;lt;math&amp;gt;&amp;lt; 0.05&amp;lt;/math&amp;gt;), all these predictors are statistically significant.&lt;br /&gt;
&lt;br /&gt;
====Example 2: Poisson Regression for Count Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with built-in R dataset: AIDS cases in Belgium&lt;br /&gt;
# Load necessary libraries&lt;br /&gt;
if (!require(&amp;quot;MASS&amp;quot;)) install.packages(&amp;quot;MASS&amp;quot;)&lt;br /&gt;
library(MASS)&lt;br /&gt;
&lt;br /&gt;
# Load AIDS dataset&lt;br /&gt;
data(Aids2)&lt;br /&gt;
cat(&amp;quot;AIDS dataset structure:\n&amp;quot;)&lt;br /&gt;
str(Aids2)&lt;br /&gt;
&lt;br /&gt;
# Prepare data: count of AIDS cases by year and state&lt;br /&gt;
aids_counts &amp;lt;- aggregate(cbind(count = 1:nrow(Aids2)) ~ age + state,&lt;br /&gt;
                         data = Aids2, FUN = length)&lt;br /&gt;
&lt;br /&gt;
# Fit Poisson regression&lt;br /&gt;
model2 &amp;lt;- glm(count ~ age + state, family = poisson, data = aids_counts)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Poisson Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model2)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Check for Overdispersion ===\n&amp;quot;)&lt;br /&gt;
# Pearson chi-square statistic&lt;br /&gt;
pearson_chi2 &amp;lt;- sum(residuals(model2, type = &amp;quot;pearson&amp;quot;)^2)&lt;br /&gt;
df_resid &amp;lt;- df.residual(model2)&lt;br /&gt;
dispersion &amp;lt;- pearson_chi2 / df_resid&lt;br /&gt;
cat(&amp;quot;Pearson χ²:&amp;quot;, pearson_chi2, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;Dispersion parameter:&amp;quot;, dispersion, &amp;quot;\n&amp;quot;)&lt;br /&gt;
cat(&amp;quot;p-value for H0: φ=1:&amp;quot;, pchisq(pearson_chi2, df_resid, lower.tail = FALSE), &amp;quot;\n&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# If overdispersed, fit quasipoisson model&lt;br /&gt;
if (dispersion &amp;gt; 1.5) {&lt;br /&gt;
  cat(&amp;quot;\n=== Fitting Quasi-Poisson Model (accounting for overdispersion) ===\n&amp;quot;)&lt;br /&gt;
  model2_qp &amp;lt;- glm(count ~ age + state, family = quasipoisson, data = aids_counts)&lt;br /&gt;
  summary(model2_qp)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Predict expected counts&lt;br /&gt;
cat(&amp;quot;\n=== Predictions for First 10 Observations ===\n&amp;quot;)&lt;br /&gt;
predictions &amp;lt;- predict(model2, type = &amp;quot;response&amp;quot;, se.fit = TRUE)&lt;br /&gt;
pred_df &amp;lt;- data.frame(&lt;br /&gt;
  Observed = aids_counts$count[1:10],&lt;br /&gt;
  Predicted = predictions$fit[1:10],&lt;br /&gt;
  SE = predictions$se.fit[1:10],&lt;br /&gt;
  Lower = predictions$fit[1:10] - 1.96 * predictions$se.fit[1:10],&lt;br /&gt;
  Upper = predictions$fit[1:10] + 1.96 * predictions$se.fit[1:10]&lt;br /&gt;
)&lt;br /&gt;
print(pred_df)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Example 3: Gamma Regression for Positive Continuous Data====&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Example with insurance claims data&lt;br /&gt;
if (!require(&amp;quot;insuranceData&amp;quot;)) install.packages(&amp;quot;insuranceData&amp;quot;)&lt;br /&gt;
library(insuranceData)&lt;br /&gt;
data(dataCar)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;Car insurance claims dataset:\n&amp;quot;)&lt;br /&gt;
str(dataCar)&lt;br /&gt;
&lt;br /&gt;
# Filter for positive claim amounts&lt;br /&gt;
claims_positive &amp;lt;- subset(dataCar, claimcst0 &amp;gt; 0)&lt;br /&gt;
&lt;br /&gt;
# Fit Gamma GLM with log link (common for monetary amounts)&lt;br /&gt;
model3 &amp;lt;- glm(claimcst0 ~ agecat + area + veh_age,&lt;br /&gt;
              family = Gamma(link = &amp;quot;log&amp;quot;),&lt;br /&gt;
              data = claims_positive)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Gamma Model Summary ===\n&amp;quot;)&lt;br /&gt;
summary(model3)&lt;br /&gt;
&lt;br /&gt;
cat(&amp;quot;\n=== Checking Gamma Model Assumptions ===\n&amp;quot;)&lt;br /&gt;
# Check residuals&lt;br /&gt;
res_gamma &amp;lt;- residuals(model3, type = &amp;quot;deviance&amp;quot;)&lt;br /&gt;
par(mfrow = c(1, 2))&lt;br /&gt;
hist(res_gamma, main = &amp;quot;Deviance Residuals&amp;quot;, xlab = &amp;quot;Residuals&amp;quot;)&lt;br /&gt;
qqnorm(res_gamma, main = &amp;quot;Q-Q Plot of Residuals&amp;quot;)&lt;br /&gt;
qqline(res_gamma)&lt;br /&gt;
&lt;br /&gt;
# Scale-location plot&lt;br /&gt;
fitted_values &amp;lt;- fitted(model3)&lt;br /&gt;
plot(fitted_values, sqrt(abs(res_gamma)),&lt;br /&gt;
     xlab = &amp;quot;Fitted Values&amp;quot;, ylab = &amp;quot;√|Deviance Residuals|&amp;quot;,&lt;br /&gt;
     main = &amp;quot;Scale-Location Plot&amp;quot;)&lt;br /&gt;
lines(lowess(fitted_values, sqrt(abs(res_gamma))), col = &amp;quot;red&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Software Implementation in R===&lt;br /&gt;
&lt;br /&gt;
Note that `family` is a function (a &amp;quot;closure&amp;quot;), not a list or an object with a \(\$family\) component. In thecode, we're passing&lt;br /&gt;
the family function itself (e.g., gaussian) to the \(run\_glm\_analysis()\) function, and later accessing \(family\$family\), &lt;br /&gt;
which doesn’t exist—because gaussian is a function, not a fitted model object.&lt;br /&gt;
In R, gaussian, binomial, poisson, etc., are functions that return family objects when called.&lt;br /&gt;
We're passing the function, not the result of calling it.&lt;br /&gt;
However, `glm()` internally calls `family()`, i.e., `gaussian()`, to get the actual family list object, which does have a \(\$family\) element.&lt;br /&gt;
Instead of checking \(family\$family\), we extract the family name from the ''fitted model'', not from the input argument.&lt;br /&gt;
Specifically, after fitting the model, we use \(model\$family\$family\), which is a character string like &amp;quot;gaussian&amp;quot;, &amp;quot;binomial&amp;quot;, etc.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
# Comprehensive GLM analysis function&lt;br /&gt;
run_glm_analysis &amp;lt;- function(formula, data, family, link = NULL) {&lt;br /&gt;
  # Fit the model&lt;br /&gt;
  if (!is.null(link)) {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family(link = link))&lt;br /&gt;
  } else {&lt;br /&gt;
    model &amp;lt;- glm(formula, data = data, family = family)&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  # Model summary&lt;br /&gt;
  cat(&amp;quot;=== MODEL SUMMARY ===\n&amp;quot;)&lt;br /&gt;
  print(summary(model))&lt;br /&gt;
  &lt;br /&gt;
  # Confidence intervals&lt;br /&gt;
  cat(&amp;quot;\n=== 95% CONFIDENCE INTERVALS ===\n&amp;quot;)&lt;br /&gt;
  print(confint(model))&lt;br /&gt;
  &lt;br /&gt;
  # Goodness-of-fit tests&lt;br /&gt;
  cat(&amp;quot;\n=== GOODNESS-OF-FIT ===\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Null Deviance:&amp;quot;, model$null.deviance, &amp;quot;on&amp;quot;, model$df.null, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;Residual Deviance:&amp;quot;, model$deviance, &amp;quot;on&amp;quot;, model$df.residual, &amp;quot;df\n&amp;quot;)&lt;br /&gt;
  cat(&amp;quot;AIC:&amp;quot;, AIC(model), &amp;quot;\n&amp;quot;)&lt;br /&gt;
  &lt;br /&gt;
  # Check for overdispersion (for Poisson and binomial families)&lt;br /&gt;
  family_name &amp;lt;- model$family$family  # ✅ Extract from fitted model&lt;br /&gt;
  if (family_name %in% c(&amp;quot;poisson&amp;quot;, &amp;quot;binomial&amp;quot;, &amp;quot;quasipoisson&amp;quot;, &amp;quot;quasibinomial&amp;quot;)) {&lt;br /&gt;
    cat(&amp;quot;\n=== OVERDISPERSION CHECK ===\n&amp;quot;)&lt;br /&gt;
    dispersion &amp;lt;- model$deviance / model$df.residual&lt;br /&gt;
    cat(&amp;quot;Dispersion parameter:&amp;quot;, round(dispersion, 4), &amp;quot;\n&amp;quot;)&lt;br /&gt;
    if (abs(dispersion - 1) &amp;gt; 0.1) {&lt;br /&gt;
      direction &amp;lt;- ifelse(dispersion &amp;gt; 1, &amp;quot;over&amp;quot;, &amp;quot;under&amp;quot;)&lt;br /&gt;
      cat(&amp;quot;Note: Significant&amp;quot;, direction, &amp;quot;dispersion detected\n&amp;quot;)&lt;br /&gt;
    }&lt;br /&gt;
  }&lt;br /&gt;
  &lt;br /&gt;
  return(model)&lt;br /&gt;
}&lt;br /&gt;
&lt;br /&gt;
# Example usage with mtcars dataset&lt;br /&gt;
cat(&amp;quot;\n\n=== EXAMPLE: GAUSSIAN GLM (equivalent to linear regression) ===\n&amp;quot;)&lt;br /&gt;
data(mtcars)&lt;br /&gt;
model_gaussian &amp;lt;- run_glm_analysis(&lt;br /&gt;
  formula = mpg ~ wt + hp + cyl,&lt;br /&gt;
  data = mtcars,&lt;br /&gt;
  family = gaussian,&lt;br /&gt;
  link = &amp;quot;identity&amp;quot;&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Compare with lm() for verification&lt;br /&gt;
cat(&amp;quot;\n=== COMPARISON WITH lm() ===\n&amp;quot;)&lt;br /&gt;
model_lm &amp;lt;- lm(mpg ~ wt + hp + cyl, data = mtcars)&lt;br /&gt;
print(summary(model_lm))&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Common GLM Families and Links in R===&lt;br /&gt;
&lt;br /&gt;
&amp;lt;center&amp;gt;&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot; style=&amp;quot;text-align:center; width:80%&amp;quot; border=&amp;quot;1&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Family !! Default Link !! Alternative Links !! Typical Use&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || Continuous, symmetric data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;binomial()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;logit&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;probit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cauchit&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;cloglog&amp;lt;/code&amp;gt; || Binary/count proportions&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;poisson()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;sqrt&amp;lt;/code&amp;gt; || Count data&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;Gamma()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, right-skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;inverse.gaussian()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;1/μ²&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;inverse&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;log&amp;lt;/code&amp;gt;, &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || Positive continuous, highly skewed&lt;br /&gt;
|-&lt;br /&gt;
| &amp;lt;code&amp;gt;quasi()&amp;lt;/code&amp;gt; || &amp;lt;code&amp;gt;identity&amp;lt;/code&amp;gt; || User-defined || Overdispersed data&lt;br /&gt;
|}&lt;br /&gt;
&amp;lt;/center&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Practical Considerations===&lt;br /&gt;
&lt;br /&gt;
* Model Selection:&lt;br /&gt;
** AIC (Akaike Information Criterion): \( \text{AIC} = -2\ell + 2p \)&lt;br /&gt;
** BIC (Bayesian Information Criterion): \( \text{BIC} = -2\ell + p\log(n) \)&lt;br /&gt;
** Cross-validation: Particularly useful for predictive performance.&lt;br /&gt;
&lt;br /&gt;
* Handling Overdispersion: For Poisson models: \( \text{Var}(Y) = \phi\mu \) where \(\phi &amp;gt; 1\) indicates overdispersion&lt;br /&gt;
Solutions:&lt;br /&gt;
** Use quasi-Poisson model&lt;br /&gt;
** Use negative binomial distribution&lt;br /&gt;
** Include random effects.&lt;br /&gt;
&lt;br /&gt;
* Zero-Inflation: For count data with excess zeros, consider:&lt;br /&gt;
** Zero-inflated Poisson (ZIP) model&lt;br /&gt;
** Zero-inflated negative binomial (ZINB) model&lt;br /&gt;
** Hurdle models.&lt;br /&gt;
&lt;br /&gt;
===Advanced Topics===&lt;br /&gt;
&lt;br /&gt;
==== Mixed Effects GLM (GLMM)====&lt;br /&gt;
&lt;br /&gt;
Extension incorporating random effects:&lt;br /&gt;
&lt;br /&gt;
&amp;lt;math&amp;gt;&lt;br /&gt;
g(E[Y|b]) = \mathbf{X}\beta + \mathbf{Z}b&lt;br /&gt;
&amp;lt;/math&amp;gt;&lt;br /&gt;
where \( b \sim N(0, \mathbf{G}) \).&lt;br /&gt;
&lt;br /&gt;
Implementation in R with &amp;lt;code&amp;gt;lme4::glmer()&amp;lt;/code&amp;gt;:&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(lme4)&lt;br /&gt;
&lt;br /&gt;
# generate random data&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
id &amp;lt;- seq(n)&lt;br /&gt;
day &amp;lt;- 1:20&lt;br /&gt;
mydata &amp;lt;- expand.grid(id = id, day = day)&lt;br /&gt;
set.seed(1)&lt;br /&gt;
trt &amp;lt;- sample(c(&amp;quot;control&amp;quot;, &amp;quot;treat&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
sex &amp;lt;- sample(c(&amp;quot;female&amp;quot;, &amp;quot;male&amp;quot;), size = n, replace = TRUE)&lt;br /&gt;
mydata$trt &amp;lt;- trt[mydata$id]&lt;br /&gt;
mydata$sex &amp;lt;- sex[mydata$id]&lt;br /&gt;
mydata &amp;lt;- mydata[order(mydata$id, mydata$day),]&lt;br /&gt;
rownames(mydata) &amp;lt;- NULL&lt;br /&gt;
head(mydata, n = 10)&lt;br /&gt;
&lt;br /&gt;
mydata$trtsex &amp;lt;- interaction(mydata$trt, mydata$sex)&lt;br /&gt;
probs &amp;lt;- c(0.40, 0.85, 0.30, 0.50)&lt;br /&gt;
names(probs) &amp;lt;- levels(mydata$trtsex)&lt;br /&gt;
mydata$p &amp;lt;- probs[mydata$trtsex]&lt;br /&gt;
&lt;br /&gt;
set.seed(3)&lt;br /&gt;
r_probs &amp;lt;- rnorm(n = n, mean = 0, sd = 0.03)&lt;br /&gt;
mydata$random_p &amp;lt;- r_probs[mydata$id]&lt;br /&gt;
mydata$p &amp;lt;- mydata$p + mydata$random_p&lt;br /&gt;
&lt;br /&gt;
# use the probabilities to generate zeroes and ones with the binom() function.&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n = nrow(mydata), size = 1, prob = mydata$p)&lt;br /&gt;
&lt;br /&gt;
# using the sim data, inspect the first few records.&lt;br /&gt;
&lt;br /&gt;
head(mydata[c(&amp;quot;id&amp;quot;, &amp;quot;day&amp;quot;, &amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;, &amp;quot;p&amp;quot;, &amp;quot;y&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
# Example with binary outcome and random intercept&lt;br /&gt;
m &amp;lt;- glmer(y ~ trt * sex + (1|id), data = mydata, family = binomial)&lt;br /&gt;
summary(m, corr = FALSE)&lt;br /&gt;
&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
====Bayesian GLM====&lt;br /&gt;
&lt;br /&gt;
Using Markov Chain Monte Carlo (MCMC) methods.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
# Bayesian logistic regression&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(y ~ day +  trt + sex + p, # predictors,&lt;br /&gt;
                        family = binomial,&lt;br /&gt;
                        data = mydata,&lt;br /&gt;
                        prior = normal(0, 2.5),&lt;br /&gt;
                        prior_intercept = normal(0, 5))&lt;br /&gt;
&lt;br /&gt;
print(bayes_model, digits = 2)&lt;br /&gt;
&lt;br /&gt;
# Coefficient Plot (Forest Plot)&lt;br /&gt;
&lt;br /&gt;
plot(bayes_model, &amp;quot;areas&amp;quot;)  # density + intervals&lt;br /&gt;
# OR more customizable:&lt;br /&gt;
library(bayesplot)&lt;br /&gt;
mcmc_intervals(bayes_model, prob = 0.95) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Distributions of Coefficients&amp;quot;)&lt;br /&gt;
&lt;br /&gt;
# Posterior Predictive Checks (PPC)&lt;br /&gt;
# Check if the model can reproduce data like the observed&lt;br /&gt;
pp_check(bayes_model, plotfun = &amp;quot;hist&amp;quot;, nreps = 100) +&lt;br /&gt;
  ggplot2::labs(title = &amp;quot;Posterior Predictive Check (Histogram of y_rep vs y)&amp;quot;)&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
Another Bayesian Experiment.&lt;br /&gt;
&lt;br /&gt;
&amp;lt;pre&amp;gt;&lt;br /&gt;
library(rstanarm)&lt;br /&gt;
set.seed(123)&lt;br /&gt;
&lt;br /&gt;
# Simulate data — explicitly make trt and sex into factors&lt;br /&gt;
n &amp;lt;- 500&lt;br /&gt;
mydata &amp;lt;- data.frame(&lt;br /&gt;
  day = sample(1:30, n, replace = TRUE),&lt;br /&gt;
  trt = factor(sample(c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;A&amp;quot;, &amp;quot;B&amp;quot;)),&lt;br /&gt;
  sex = factor(sample(c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;), n, replace = TRUE), levels = c(&amp;quot;F&amp;quot;, &amp;quot;M&amp;quot;)),&lt;br /&gt;
  p   = rnorm(n, mean = 50, sd = 10)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# True log-odds&lt;br /&gt;
logit_p &amp;lt;- -1 - 0.02 * mydata$day + &lt;br /&gt;
           0.6 * (as.numeric(mydata$trt) - 1) +   # A=0, B=1&lt;br /&gt;
           0.3 * (as.numeric(mydata$sex) - 1) +    # F=0, M=1&lt;br /&gt;
           0.04 * mydata$p&lt;br /&gt;
&lt;br /&gt;
mydata$y &amp;lt;- rbinom(n, size = 1, prob = plogis(logit_p))&lt;br /&gt;
&lt;br /&gt;
# Fit model&lt;br /&gt;
bayes_model &amp;lt;- stan_glm(&lt;br /&gt;
  y ~ day + trt + sex + p,&lt;br /&gt;
  family = binomial,&lt;br /&gt;
  data = mydata,&lt;br /&gt;
  prior = normal(0, 2.5),&lt;br /&gt;
  prior_intercept = normal(0, 5),&lt;br /&gt;
  seed = 42,&lt;br /&gt;
  refresh = 0&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Create New Data&lt;br /&gt;
# Confirm variables are factors&lt;br /&gt;
str(mydata[c(&amp;quot;trt&amp;quot;, &amp;quot;sex&amp;quot;)])&lt;br /&gt;
&lt;br /&gt;
# Create prediction grid&lt;br /&gt;
newdata &amp;lt;- expand.grid(&lt;br /&gt;
  day = seq(min(mydata$day), max(mydata$day), length.out = 30),&lt;br /&gt;
  trt = levels(mydata$trt)[1],      # e.g., &amp;quot;A&amp;quot;&lt;br /&gt;
  sex = levels(mydata$sex)[1],      # e.g., &amp;quot;F&amp;quot;&lt;br /&gt;
  p   = median(mydata$p)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Verify it worked&lt;br /&gt;
stopifnot(nrow(newdata) &amp;gt; 0)&lt;br /&gt;
str(newdata)&lt;br /&gt;
&lt;br /&gt;
# Generate Predictions from the Bayesian Posterior Probability (after fitting the Bayesian Model):&lt;br /&gt;
# Posterior predicted probabilities&lt;br /&gt;
post_pred &amp;lt;- posterior_linpred(bayes_model, newdata = newdata, transform = TRUE)&lt;br /&gt;
&lt;br /&gt;
# Summarize&lt;br /&gt;
pred_summary &amp;lt;- data.frame(&lt;br /&gt;
  day = newdata$day,&lt;br /&gt;
  prob = apply(post_pred, 2, median),&lt;br /&gt;
  lower = apply(post_pred, 2, quantile, 0.025),&lt;br /&gt;
  upper = apply(post_pred, 2, quantile, 0.975)&lt;br /&gt;
)&lt;br /&gt;
&lt;br /&gt;
# Plot&lt;br /&gt;
library(ggplot2)&lt;br /&gt;
ggplot(pred_summary, aes(x = day, y = prob)) +&lt;br /&gt;
  geom_line(color = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  geom_ribbon(aes(ymin = lower, ymax = upper), alpha = 0.2, fill = &amp;quot;blue&amp;quot;) +&lt;br /&gt;
  labs(&lt;br /&gt;
    title = &amp;quot;Predicted Probability of Outcome by Day&amp;quot;,&lt;br /&gt;
    subtitle = &amp;quot;Holding treatment = A, sex = F, and p = median&amp;quot;,&lt;br /&gt;
    y = &amp;quot;P(y = 1)&amp;quot;,&lt;br /&gt;
    x = &amp;quot;Day&amp;quot;&lt;br /&gt;
  ) +&lt;br /&gt;
  theme_minimal()&lt;br /&gt;
&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===Problems===&lt;br /&gt;
&lt;br /&gt;
1. Conceptual Exercises:&lt;br /&gt;
&lt;br /&gt;
   a) Derive the score equations for a Poisson GLM with log link&lt;br /&gt;
   b) Show that the binomial distribution with logit link is the canonical link&lt;br /&gt;
   c) Prove that the deviance for a normal GLM equals the residual sum of squares&lt;br /&gt;
&lt;br /&gt;
2. Applied Problems:&lt;br /&gt;
&lt;br /&gt;
   a) Analyze the [http://wiki.socr.umich.edu/index.php/SOCR_Data_Dinov_021808_ConsumerPriceIndex3Way Consumer Price Index] data using appropriate GLM&lt;br /&gt;
   b) Model the [http://wiki.socr.umich.edu/index.php/SOCR_Data_MonetaryBase1959_2009 Monetary Base] data considering temporal autocorrelation&lt;br /&gt;
   c) Using the &amp;lt;code&amp;gt;iris&amp;lt;/code&amp;gt; dataset, build a multinomial logistic regression to classify species&lt;br /&gt;
&lt;br /&gt;
3. Practice Simulation Study:&lt;br /&gt;
&lt;br /&gt;
   &amp;lt;pre&amp;gt;&lt;br /&gt;
   # Simulate data from a logistic regression model&lt;br /&gt;
   set.seed(123)&lt;br /&gt;
   n &amp;lt;- 1000&lt;br /&gt;
   x1 &amp;lt;- rnorm(n)&lt;br /&gt;
   x2 &amp;lt;- rnorm(n)&lt;br /&gt;
   beta &amp;lt;- c(0.5, 1, -0.5)&lt;br /&gt;
   linear_predictor &amp;lt;- beta[1] + beta[2]*x1 + beta[3]*x2&lt;br /&gt;
   probabilities &amp;lt;- plogis(linear_predictor)&lt;br /&gt;
   y &amp;lt;- rbinom(n, size = 1, prob = probabilities)&lt;br /&gt;
   &lt;br /&gt;
   # Fit model and evaluate performance&lt;br /&gt;
   sim_data &amp;lt;- data.frame(y = y, x1 = x1, x2 = x2)&lt;br /&gt;
   model_sim &amp;lt;- glm(y ~ x1 + x2, family = binomial, data = sim_data)&lt;br /&gt;
   &lt;br /&gt;
   # Calculate bias and MSE&lt;br /&gt;
   beta_hat &amp;lt;- coef(model_sim)&lt;br /&gt;
   bias &amp;lt;- beta_hat - beta&lt;br /&gt;
   mse &amp;lt;- mean((beta_hat - beta)^2)&lt;br /&gt;
   &lt;br /&gt;
   cat(&amp;quot;True parameters:&amp;quot;, beta, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Estimated parameters:&amp;quot;, beta_hat, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;Bias:&amp;quot;, bias, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   cat(&amp;quot;MSE:&amp;quot;, mse, &amp;quot;\n&amp;quot;)&lt;br /&gt;
   &amp;lt;/pre&amp;gt;&lt;br /&gt;
&lt;br /&gt;
===References===&lt;br /&gt;
&lt;br /&gt;
1. McCullagh, P., &amp;amp; Nelder, J. A. (1989). *Generalized Linear Models* (2nd ed.). Chapman and Hall.&lt;br /&gt;
&lt;br /&gt;
2. Dobson, A. J., &amp;amp; Barnett, A. G. (2018). *An Introduction to Generalized Linear Models* (4th ed.). CRC Press.&lt;br /&gt;
&lt;br /&gt;
3. Agresti, A. (2015). *Foundations of Linear and Generalized Linear Models*. Wiley.&lt;br /&gt;
&lt;br /&gt;
4. Fahrmeir, L., Kneib, T., Lang, S., &amp;amp; Marx, B. (2013). *Regression: Models, Methods and Applications*. Springer.&lt;br /&gt;
&lt;br /&gt;
5. Wood, S. N. (2017). *Generalized Additive Models: An Introduction with R* (2nd ed.). Chapman and Hall/CRC.&lt;br /&gt;
&lt;br /&gt;
====Online Resources====&lt;br /&gt;
* [https://en.wikipedia.org/wiki/Generalized_linear_model GLM Wikipedia]&lt;br /&gt;
* [https://data.princeton.edu/wws509/notes/a2.pdf GLM Theory Notes]&lt;br /&gt;
* [https://www.jstatsoft.org/article/view/v015i12 GLM in R Tutorial]&lt;br /&gt;
* [https://cran.r-project.org/web/views/SocialSciences.html R Resources for Social Sciences]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
* SOCR Home page: https://www.socr.umich.edu&lt;br /&gt;
&lt;br /&gt;
{{translate|pageName=http://wiki.socr.umich.edu/index.php?title=SMHS_GLM}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18350</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18350"/>
		<updated>2026-03-09T01:07:31Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Course Coverage */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
: A ''heuristic overview'' of complex-time representation, kime-phase tomography, and spacekime analytics.&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/Spacekime_TCIU_Podcast1_2025.mp3 A brief, 7-minute, spacekime ''podcast'' (audio)].&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/ComplexTimeReps_KPT_Spacekime_FieldEquations_video.mp4 A short, 5-minute ''video'' describing complex-time representation, kime-phase tomography (KPT), and spacekime gravitational field equations].&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data.&lt;br /&gt;
:: [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Kime-Phase Tomography (KPT) (slidedeck)], &lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves],&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
: [https://tciu.predictive.space/ Additional theoretical developments, experimental demonstrations, computational algorithms, and data science applications are available on the '''spacekime/TCIU website'''].&lt;br /&gt;
&lt;br /&gt;
: [https://socr.umich.edu/DSPA2/DSPA2_notes/01_Introduction.html#2_Foundations_of_R More information about editing, interpreting, knitting/compiling and running ''R''-markdown electronic notebooks (*.Rmd) is available on the SOCR DSPA2 site].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18349</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18349"/>
		<updated>2026-03-09T00:46:22Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Course Coverage */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
: A ''heuristic overview'' of complex-time representation, kime-phase tomography, and spacekime analytics.&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/Spacekime_TCIU_Podcast1_2025.mp3 A brief, 7-minute, spacekime ''podcast'' (audio)].&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/ComplexTimeReps_KPT_Spacekime_FieldEquations_video.mp4 A short, 5-minute ''video'' describing complex-time representation, kime-phase tomography (KPT), and spacekime gravitational field equations].&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data.&lt;br /&gt;
:: [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Kime-Phase Tomography (KPT) (slidedeck)], &lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves],&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
: [https://tciu.predictive.space/ Additional theoretical developments, experimental demonstrations, computational algorithms, and data science applications are available on the '''spacekime/TCIU website'''].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18348</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18348"/>
		<updated>2026-03-09T00:44:47Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Description */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
: A ''heuristic overview'' of complex-time representation, kime-phase tomography, and spacekime analytics.&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/Spacekime_TCIU_Podcast1_2025.mp3 A brief, 7-minute, spacekime ''podcast'' (audio)].&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/ComplexTimeReps_KPT_Spacekime_FieldEquations_video.mp4 A short, 5-minute ''video'' describing complex-time representation, kime-phase tomography (KPT), and spacekime gravitational field equations].&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data.&lt;br /&gt;
:: [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part I.1: Kime-Phase Tomography (KPT) (slidedeck)], &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Part I.2: Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves],&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html Part I.3: 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
:: ...&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
: [https://tciu.predictive.space/ Additional theoretical developments, experimental demonstrations, computational algorithms, and data science applications are available on the '''spacekime/TCIU website'''].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18347</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18347"/>
		<updated>2026-03-09T00:43:20Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Description */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
: Heuristic overview of complex-time representation, kime-phase tomography, and spacekime analytics:&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/Spacekime_TCIU_Podcast1_2025.mp3 A brief, 7-minute, spacekime ''podcast'' (audio)].&lt;br /&gt;
:: [https://socr.umich.edu/spacekime/img/ComplexTimeReps_KPT_Spacekime_FieldEquations_video.mp4 A short, 5-minute ''video'' describing complex-time representation, kime-phase tomography (KPT), and spacekime gravitational field equations].&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data.&lt;br /&gt;
:: [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part I.1: Kime-Phase Tomography (KPT) (slidedeck)], &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Part I.2: Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves],&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html Part I.3: 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
:: ...&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
: [https://tciu.predictive.space/ Additional theoretical developments, experimental demonstrations, computational algorithms, and data science applications are available on the '''spacekime/TCIU website'''].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18346</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18346"/>
		<updated>2026-03-08T22:19:21Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Course Coverage */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data.&lt;br /&gt;
:: [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part I.1: Kime-Phase Tomography (KPT) (slidedeck)], &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Part I.2: Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves],&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html Part I.3: 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
:: ...&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
: [https://tciu.predictive.space/ Additional theoretical developments, experimental demonstrations, computational algorithms, and data science applications are available on the '''spacekime/TCIU website'''].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18345</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18345"/>
		<updated>2026-03-08T22:16:55Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Course Coverage */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data.&lt;br /&gt;
:: [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part I.1: Kime-Phase Tomography (KPT) (slidedeck)], &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Part I.2: Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves],&lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html Part I.3: 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
:: ...&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. &lt;br /&gt;
:: [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18344</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18344"/>
		<updated>2026-03-08T22:16:02Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Course Coverage */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data, [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part I.1: Kime-Phase Tomography (KPT) (slidedeck)], [https://www.socr.umich.edu/TCIU/HTMLs/Kepler_KPT_Implementation_V5.html Part I.2: Kime-Phase Tomography: (V.5) KPT Validation with Kepler Stellar Light Curves], and [https://www.socr.umich.edu/TCIU/HTMLs/TCIU_SK_Appendix04_Spacekime_Gravity_FieldEquations_V2.html Part I.3: 5D Spacekime Einstein-Hilbert Gravitational Field Equations].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_MappingLongitudinalTimeseries_2_Kimesurfaces.html Strategies for Mapping Repeated Measurement Longitudinal Data (Time-series) to 2D Manifolds (Kimesurfaces)].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics. [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18343</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18343"/>
		<updated>2026-03-08T22:11:16Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Course Coverage */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data, [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part 1: Kime-Phase Tomography (KPT) (slidedeck)].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference. [https://www.socr.umich.edu/TCIU/HTMLs/Chapter6_TCIU_Basic_SpacekimePredictiveAnalytics.html TCIU/Spacekime Analytics Tutorial: Basic TCIU Protocol for Predictive Spacekime Analytics using Longitudinal Data].&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18342</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18342"/>
		<updated>2026-03-08T15:10:09Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Sessions */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data, [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part 1: Kime-Phase Tomography (KPT) (slidedeck)].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference.&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: Exploring the Relevance of Thermal Spacetime Quantum Mechanics to Data Science and Medical Imaging&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18341</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18341"/>
		<updated>2026-03-08T15:08:06Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Sessions */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data, [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part 1: Kime-Phase Tomography (KPT) (slidedeck)].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference.&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''PDF Agenda''': [https://socr.umich.edu/docs/uploads/2026/APS_GPS_Mar-J57_Session_Agenda.pdf MAR-J57 Session Program (PDF)]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: [Title TBD]&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
	<entry>
		<id>https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18340</id>
		<title>SOCR News APS GPS 2026</title>
		<link rel="alternate" type="text/html" href="https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026&amp;diff=18340"/>
		<updated>2026-03-08T15:01:21Z</updated>

		<summary type="html">&lt;p&gt;Dinov: /* Course Coverage */&lt;/p&gt;
&lt;hr /&gt;
&lt;div&gt;== [[SOCR_News | SOCR News &amp;amp; Events]]: '''2026 APS Global Physics Summit (GPS)''' ==&lt;br /&gt;
&lt;br /&gt;
==2026 APS Global Physics Summit==&lt;br /&gt;
&lt;br /&gt;
[[Image:APS_2025_Summit.png|500px|thumbnail|center| [https://summit.aps.org/ 2026 APS Global Summit] ]]&lt;br /&gt;
&lt;br /&gt;
The [https://meetings.aps.org annual American Physical Society's March Meetings] are scientific research conferences convening over 14,000 physicists and students from around the globe to connect and collaborate across academia, industry, and major labs. The [https://summit.aps.org/ 2026 '''APS Global Physics Summit'''] will take place at [https://denverconvention.com/ Colorado Convention Center, Denver Colorado]. The [https://summit.aps.org/schedule/ APS 2026 Summit Program is here].&lt;br /&gt;
&lt;br /&gt;
The SOCR group is organizing a Time-Complexity and Inferential Uncertainty (TCIU), Spacekime Analytics, short course, organize sessions, and present invited talks.&lt;br /&gt;
&lt;br /&gt;
===[https://summit.aps.org/schedule/ APS GPS 2026 Conference Program/Schedule]===&lt;br /&gt;
&lt;br /&gt;
==Short Course (Tutorial) ==&lt;br /&gt;
&lt;br /&gt;
===Logistics===&lt;br /&gt;
&lt;br /&gt;
* ''Event'': [https://summit.aps.org/smt/2026/events/MAR-SH03 Complex-Time (Kime) Representation of Spatiotemporal Processes and Spacekime Analytics]&lt;br /&gt;
* ''Sponsor'': [https://engage.aps.org/dcomp/home APS Division of Computational Physics (DCOMP)]&lt;br /&gt;
* ''Date/Time'': Sunday, March 15, 2026, Time: 8:00am-12:00pm&lt;br /&gt;
* ''Place'': [https://denverconvention.com/ Colorado Convention Center, Room 702]&lt;br /&gt;
* ''Price'': Students: $10; Postdocs/fellows: $25; Academic faculty: $50; Federal Employees: $50; Industry: $75; International Participants: $30.&lt;br /&gt;
&lt;br /&gt;
* ''Who should attend?'' This short course may be appropriate for undergraduate seniors, graduate students, postdocs and fellows, research scientists, and early career faculty interested in mathematical representation, computational modeling, statistical inference, and AI. While all content will be broadly accessible, a background in mathematical physics and a strong interest in advanced statistical learning methods and AI applications will be beneficial. Participants motivated to explore the intersection of statistical physics, computing, and AI will gain the most from learning spacekime analytics and completing this course. Additionally, some knowledge of statistical limit theory, functional analysis, and experience with R/Rmd programming may be helpful. All interested students, fellows, early-career faculty, and members of [https://engage.aps.org/gds/home GDS] and [https://engage.aps.org/dcomp/home DCOMP] are encouraged to attend.&lt;br /&gt;
&lt;br /&gt;
* ''Organizer'': [https://socr.umich.edu/people/dinov/ Ivo D. Dinov]&lt;br /&gt;
&lt;br /&gt;
===Description===&lt;br /&gt;
* ''Description'': Fundamentally new representations of complex systems are critical in advancing scientific discovery, transdisciplinary research, and blended data-driven human-machine intelligence. This PEP course will present the fundamentals of complex-time (kime) representation, which interfaces quantum mechanics, mathematical statistics, and AI applications. Specifically, the course will demonstrate computational inference, scientific visualization, rigorous statistical modeling of large datasets, understanding complex temporally dynamic processes, and spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
: The course will provide an active learning environment to build competency in the following areas: methods for transforming time-series to kime-surfaces, which are much richer computational objects; mathematical foundations of spacekime analytics; and AI algorithms with applications. A set of end-to-end R-markdown notebooks will provide reproducible and independently validated protocols for data ingestion, mathematical representation and scientific visualization of longitudinal time-courses mapped as 2D parametric manifolds (kime-surfaces), model-based statistical inference, and model-free AI analytics.&lt;br /&gt;
&lt;br /&gt;
===Course Coverage===&lt;br /&gt;
&lt;br /&gt;
* '''Part I (8:00-8:55 AM)''': Course overview with a discussion of the basic mathematical-physics principles underlying the complex-time (''kime'') representation of repeated measurement longitudinal data, [https://socr.umich.edu/docs/uploads/2026/APS_GPS__2026_KPT_long_Dinov_Tutorial_P1.pdf Part 1: Kime-Phase Tomography (KPT) (slidedeck)].&lt;br /&gt;
&lt;br /&gt;
* '''Part II (9:00-9:55 AM)''': Translation of fundamental quantum mechanics principles into statistical inference models of time-varying processes. Specifically, we will demonstrate generalizing the classical 4D spatiotemporal sampling to a 5D spacekime manifold, where the phase of complex-time (an extra degree of freedom) encodes repeated random sampling at fixed spatiotemporal locations.&lt;br /&gt;
&lt;br /&gt;
* '''Part III (10:00-10:55 AM)''': Alternative strategies for transforming observed time-series into kime-surfaces, which are richer, computationally tractable, objects amenable to tensor-based statistical linear modeling and AI model-free inference.&lt;br /&gt;
&lt;br /&gt;
* '''Part IV (11:00-11:55 AM)''': Simulated and real neuroimaging and macroeconomics data to demonstrate applications of spacekime analytics.&lt;br /&gt;
&lt;br /&gt;
===Spacekime analytics themes===&lt;br /&gt;
&lt;br /&gt;
* Importing of repeated measurement longitudinal data,&lt;br /&gt;
&lt;br /&gt;
* Numeric (stitching) and analytic (Laplace) kimesurface reconstruction from time-series data,&lt;br /&gt;
&lt;br /&gt;
* Forward prediction modeling extrapolating the process behavior beyond the observed time-span&lt;br /&gt;
&lt;br /&gt;
* Group comparison discrimination between cohorts based on the structure and properties of their corresponding kimesurfaces. For instance, statistically quantify the differences between two or more groups,&lt;br /&gt;
&lt;br /&gt;
*  Unsupervised clustering and classification of individuals, traits, and other latent characteristics of cases included in the study,&lt;br /&gt;
&lt;br /&gt;
* Constructing low-dimensional visual representations of large repeated measurement data across multiple individuals as pooled kimesurfaces (parameterized 2D manifolds),&lt;br /&gt;
&lt;br /&gt;
* Statistical comparison, topological quantification, and analytical inference using kimesurface representations of repeated-measurement longitudinal data.&lt;br /&gt;
&lt;br /&gt;
===Expected Learning Outcomes===&lt;br /&gt;
&lt;br /&gt;
Participants will learn and understand the rationale for, and the duality between, classical representation of repeated-measurement spacetime observations (time-series) and their spacekime analytics counterparts – complex-time representation (parametric manifolds).&lt;br /&gt;
&lt;br /&gt;
== Sessions ==&lt;br /&gt;
&lt;br /&gt;
===Session [https://summit.aps.org/smt/2026/events/MAR-J57 MAR-J57: Physical, Statistical and Al modeling of Spatiotemporal Medical Imaging]===&lt;br /&gt;
&lt;br /&gt;
* '''Focus Session''': [https://summit.aps.org/smt/2026/events/MAR-J57 Session MAR-J57]&lt;br /&gt;
* '''Sponsoring Unit''': [https://engage.aps.org/gmed/home GMED]&lt;br /&gt;
* '''Date/Time''': Tuesday, March 17, 2026, 3:30-6:00 PM CT&lt;br /&gt;
* '''Room''': [https://denverconvention.com/plan-your-event/event-space/bluebird-ballroom Convention Center Bluebird 2A]&lt;br /&gt;
* '''Session Chairs''': [https://www.socr.umich.edu/people/dinov/ Ivo Dinov] &amp;amp; [https://radiationoncology.pitt.edu/people/sarah-ashmeg-phd-dabr Sarah Ashmeg]&lt;br /&gt;
&lt;br /&gt;
{| class=&amp;quot;wikitable&amp;quot;&lt;br /&gt;
|-&lt;br /&gt;
! Time&lt;br /&gt;
! Presentation&lt;br /&gt;
! Speaker&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;3:30 PM – 4:06 PM&lt;br /&gt;
| MAR-J57.00001: AMPERE: A Global Physics Framework for Advancing Medical Physics Research, Education, and Workforce Development&lt;br /&gt;
| Stephen M Avery (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:06 PM – 4:42 PM&lt;br /&gt;
| MAR-J57.00002: Complex Time and the Principle of Least Entropic Purpose&lt;br /&gt;
| Michael C Parker (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;4:42 PM – 5:18 PM&lt;br /&gt;
| MAR-J57.00003: [Title TBD]&lt;br /&gt;
| Russell Thompson (Invited)&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:18 PM – 5:30 PM&lt;br /&gt;
| MAR-J57.00004: Second-order Statistical Image Texture Features from Spectral X-ray Images when using Photon-Counting Spectral Detectors&lt;br /&gt;
| Oliver Namuwonge, Diego Andrade, Mini Das&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:30 PM – 5:42 PM&lt;br /&gt;
| MAR-J57.00005: Fractal Diffusion-Reaction Framework for FLASH-UHDR Tissue Sparing&lt;br /&gt;
| Neda Valizadeh Gendeshmin, Robabeh Rahimi, Ramin Abolfazl&lt;br /&gt;
|-&lt;br /&gt;
| Tuesday, March 17, 2026&amp;lt;br /&amp;gt;5:42 PM – 5:54 PM&lt;br /&gt;
| MAR-J57.00006: Estimating Uncertainty of a Deep Learning-Based Breast Cancer Risk Prediction Model Using Test-Time Augmentations&lt;br /&gt;
| Zan Klanecek, Yao Kuan Wang, Tobias Wagner, Lesley Cockmartin, Nicholas Marshall, Andrej Studen, Katja Jarm, Mateja Krajc, Milos Vrhovec, Hilde Bosmans, Ali Deatsch, Robert Jeraj&lt;br /&gt;
|}&lt;br /&gt;
&lt;br /&gt;
==Talks==&lt;br /&gt;
&lt;br /&gt;
* '''Session''': [https://summit.aps.org/smt/2026/events/MAR-J45 MAR-J45], time: 04:54p - 05:06p&lt;br /&gt;
* '''Talk Title''': [https://summit.aps.org/events/MAR-J45/6 Kime-Phase Tomography and Manifold Representation of Longitudinal Processes over a Complex-Time Domain]&lt;br /&gt;
* '''Presenter''': [https://www.socr.umich.edu/people/dinov/ Ivo D. Dinov] (presenter), Yueyang Shen, Bojko N Bakalov&lt;br /&gt;
* '''Slides''': [https://socr.umich.edu/docs/uploads/2026/Dinov_short_ID_APS_GPS__2026_KPT.pdf Dinov's (short_ID_APS_GPS__2026_KPT) Slidedeck (PDF)]&lt;br /&gt;
* '''Session Logistics''': Tue. March 17, 03:30p - 06:06p, Convention Center, Meeting Room 605&lt;br /&gt;
&lt;br /&gt;
==Other Sessions==&lt;br /&gt;
&lt;br /&gt;
* Sunday 3/15 at 6:00pm (location: TBD): GDS (off-site) executive committee meeting&lt;br /&gt;
* TBD: GDS GPS general meeting.&lt;br /&gt;
&lt;br /&gt;
==Resources==&lt;br /&gt;
* [https://wiki.socr.umich.edu/images/b/ba/Dinov_Spacekime_2025_Slidedeck_APS_Summit.pdf Dinov's ''Numerical and Analytical Complex-time Transformations of Longitudinal Processes and Spacekime Analytics'' slides]&lt;br /&gt;
* [https://socr.umich.edu/docs/uploads/2025/Shen_APS_2025.pdf Shen's ''Probabilistic Symmetry, Variable Exchangeability, and Deep Network Learning Invariance and Equivariance'' slides]&lt;br /&gt;
&lt;br /&gt;
&amp;lt;hr&amp;gt;&lt;br /&gt;
{{translate|pageName=https://wiki.socr.umich.edu/index.php?title=SOCR_News_APS_GPS_2026}}&lt;/div&gt;</summary>
		<author><name>Dinov</name></author>
		
	</entry>
</feed>