Simple Linear Regression

/*

July 2006. Annie Che <chea@stat.ucla.edu>. UCLA Statistics.

Source of example data:
An Introduction to Computational Statitics by Robert I. Jennrich.
Page 5, example of regression on students' midterm and final scores.

*/
package edu.ucla.stat.SOCR.analyses.example;

import java.util.HashMap;
import edu.ucla.stat.SOCR.analyses.data.Data;
import edu.ucla.stat.SOCR.analyses.data.DataType;
import edu.ucla.stat.SOCR.analyses.result.SimpleLinearRegressionResult;


public class SimpleLinearRegressionExample {
	public static void main(String args[]) {
		double[] midtermGrade =            
                {68,49,60,68,97,82,59,50,73,39,71,95,61,72,87,40,66,58,58,77};
		double[] finalGrade = 
               {75,63,57,88,88,79,82,73,90,62,70,96,76,75,85,40,74,70,75,72};

		// you'll need to instantiate a data instance first.
		Data data = new Data();

		/*********************************************************************
		then put the data into the Data Object.
		append the predictor data using method "addPredictor".
		append the response data using method "addResponse".
		**********************************************************************/

		data.addPredictor(midtermGrade, DataType.QUANTITATIVE);
		data.addResponse(finalGrade, DataType.QUANTITATIVE);


		try {
			SimpleLinearRegressionResult result = data.modelSimpleLinearRegression();
			if (result != null) {

				// Getting the model's parameter estiamtes and statistics.

				double alpha = result.getAlpha();
				double beta = result.getBeta();

				double meanX = result.getMeanX();
				double meanY = result.getMeanY();

				double seAlpha = result.getAlphaSE();
				double seBeta = result.getBetaSE();
				double tStatAlpha = result.getAlphaTStat();
				double tStatBeta = result.getBetaTStat();

				/* to avoid cases such "p-value < 0.0001" sometimes generated
                                by R, String is used for p-values. */

				String pvAlpha = result.getAlphaPValue();
				String pvBeta = result.getBetaPValue();

				double[] predicted = result.getPredicted();
				double[] residuals = result.getResiduals();

				// residuals after being sorted ascendantly.
				double[] sortedResiduals = result.getSortedResiduals();

				// sortedResiduals after being standardized.
				double[] sortedStandardizedResiduals =  
                                         result.getSortedStandardizedResiduals();

				// the original index of sortedResiduals, stored as integer array.
				int[] sortedResidualsIndex = result.getSortedResidualsIndex();

				// the normal quantiles of sortedResiduals.
				double[] sortedNormalQuantiles = result.getSortedNormalQuantiles();

	                			// sortedNormalQuantiles after being standardized.
				double[] sortedStandardizedNormalQuantiles = 
                                         result.getSortedStandardizedNormalQuantiles();

				System.out.println("intercept = " + alpha);
				System.out.println("slope = " + beta);
				System.out.println("meanX = " + meanX);
				System.out.println("meanY = " + meanY);

				System.out.println("seAlpha = " + seAlpha);
				System.out.println("seBeta = " + seBeta);
				System.out.println("tStatAlpha = " + tStatAlpha);
				System.out.println("tStatBeta = " + tStatBeta);

				System.out.println("pvAlpha = " + pvAlpha);
				System.out.println("pvBeta = " + pvBeta);
				for (int i = 0; i < residuals.length; i++) {
					System.out.println("residuals["+i+"] = " + residuals[i]);
				}

			}
		} catch (Exception e) {
			System.out.println(e);
		}
	}
}