SMHS BigDataBigSci GCM - Revision history

Pineaumi: /* See also */

2016-05-23T20:00:01Z

‎See also

2016-05-23T19:58:06Z

‎Model-based Analytics - Growth Curve Models

2016-05-23T19:43:57Z

‎Footnotes

2016-05-23T19:14:35Z

‎Footnotes

2016-05-23T19:14:10Z

‎Footnotes

2016-05-23T19:12:55Z

‎Comparative Fit Index

2016-05-23T13:44:02Z

‎Appendix

2016-05-23T13:43:31Z

‎Respiratory Illness GEE R example

2016-05-23T13:42:45Z

‎Root Mean Square Error of Approximation

2016-05-23T13:42:13Z

‎Comparative Fit Index

@@ Line 511: / Line 511: @@
 * [[SMHS_BigDataBigSci| Back to Model-based Analytics]]
 * [[SMHS_BigDataBigSci_SEM| Structural Equation Modeling (SEM)]]
-* [[SMHS_BigDataBigSci_GEE| Next Section: Generalized Estimating Equation (GEE) Modeling]]
 <hr>

@@ Line 76: / Line 76: @@
   # the z-values, the standardized parameter values, and returns a data frame
   fitted(fit4)	# return the model-implied (fitted) covariance matrix (and mean vector) of a fitted model
   # resid() function return (unstandardized) residuals of a fitted model including the difference between

← Older revision		Revision as of 19:43, 23 May 2016
Line 506:		Line 506:

	*<sup>2</sup> http://www.imachordata.com/ecological-sems-and-composite-variables-what-why-and-how/		*<sup>2</sup> http://www.imachordata.com/ecological-sems-and-composite-variables-what-why-and-how/
		+	*<sup>3</sup> http://www.jstatsoft.org/v15/i02/
		+	*<sup>4</sup> https://books.google.com/books?id=mdEqBgAAQBAJ

	==See also==		==See also==

← Older revision		Revision as of 19:14, 23 May 2016
Line 506:		Line 506:

	*<sup>2</sup> http://www.imachordata.com/ecological-sems-and-composite-variables-what-why-and-how/		*<sup>2</sup> http://www.imachordata.com/ecological-sems-and-composite-variables-what-why-and-how/
−
−	*<sup>3</sup> http://www.jstatsoft.org/v15/i02/
−
−	*<sup>4</sup> https://books.google.com/books?id=mdEqBgAAQBAJ

	==See also==		==See also==

@@ Line 505: / Line 505: @@
 ===Footnotes===
-* <sup>3</sup> http://www.jstatsoft.org/v15/i02/
+*<sup>2</sup> http://www.imachordata.com/ecological-sems-and-composite-variables-what-why-and-how/
-* <sup>4</sup> https://books.google.com/books?id=mdEqBgAAQBAJ
+*<sup>3</sup> http://www.jstatsoft.org/v15/i02/
 ==See also==

@@ Line 91: / Line 91: @@
 (CFI) is an incremental measure directly based on the non-centrality measure.  If d = χ2(df) where df are the degrees of freedom of the model, the Comparative Fit Index is:
-\begin{equation}
 \frac{(Null Model)-d(Proposed Model)}{d(Null Model)}.
-\end{equation}
 $0≤CFI≤1$ (by definition). It is interpreted as:

@@ Line 495: / Line 495: @@
 SEM References
-<li>http://socserv.mcmaster.ca/jfox/Misc/sem/SEM-paper.pdf </li>
+*http://socserv.mcmaster.ca/jfox/Misc/sem/SEM-paper.pdf
 GEE References
-<li>https://cran.r-project.org/web/packages/geepack/geepack.pdf</li>
+*https://cran.r-project.org/web/packages/geepack/geepack.pdf
-<li>http://www.jstatsoft.org/v15/i02/paper</li>
+*http://www.jstatsoft.org/v15/i02/paper
 ===Footnotes===

@@ Line 421: / Line 421: @@
   # GEE modeling: R function arguments/options
-<li><b>corstr</b>= for defining the correlation structure within groups in a GEE model</li>
+*<b>corstr</b>= for defining the correlation structure within groups in a GEE model
-<li><b>id</b>= is used to identify the grouping variable in a GEE model</li>
+*<b>id</b>= is used to identify the grouping variable in a GEE model
-<li><b>scale.fix</b>= when TRUE causes the scale parameter to be fixed (by default at 1) rather than estimated</li>
+*<b>scale.fix</b>= when TRUE causes the scale parameter to be fixed (by default at 1) rather than estimated
-<li><b>waves</b>= names a positive integer-valued variable that is used to identify the order and spacing of observations within groups in a GEE model. This argument is crucial when there are missing values and gaps in the data</li>
+*<b>waves</b>= names a positive integer-valued variable that is used to identify the order and spacing of observations within groups in a GEE model. This argument is crucial when there are missing values and gaps in the data
   gee.model1 <- <b>geeglm</b>(outcome ~ center + treat + sex + baseline + age, data = respiratory, family = "binomial", id = id, corstr = "exchangeable", scale.fix = TRUE)

@@ Line 114: / Line 114: @@
 where N the sample size and df the degrees of freedom of the model.  If χ<sup>2</sup>  < df, then the RMSEA∶=0.  It has a penalty for complexity via the chi square to df ratio.  The RMSEA is a popular measure of model fit.
-<li>RMSEA < 0.01, excellent, </li>
+*RMSEA < 0.01, excellent,
-<li>RMSEA  < 0.05, good </li>
+*RMSEA  < 0.05, good
-<li>RMSEA > 0.10 cutoff for poor fitting models</li>
+*RMSEA > 0.10 cutoff for poor fitting models
 ====Standardized Root Mean Square Residual====

@@ Line 97: / Line 97: @@
 $0≤CFI≤1$ (by definition). It is interpreted as:
-<li>$CFI<0.9$  - model fitting is poor. </li>
+*$CFI<0.9$  - model fitting is poor.
-<li>$0.9≤CFI≤0.95$  is considered marginal, </li>
+*$0.9≤CFI≤0.95$  is considered marginal,
-<li>$CFI>0.95$   is good.  </li>
+*$CFI>0.95$   is good.
 CFI is a relative index of model fit – it compare the fit of your model to the fit of (the worst) fitting null model.