Package gorsat.Regression

Class LogisticRegression

java.lang.Object
gorsat.Regression.LogisticRegression
public abstract class LogisticRegression extends Object
This program fits a logistic regression model on a given data set. We find the maximum likelihood estimators by using Newton's method to find the zero of the gradient of the likelihood function.

  • Field Details

    • numberOfDepVars

      protected final int numberOfDepVars

    • maximumNumberOfSamples

      protected final int maximumNumberOfSamples

    • dim

      protected final int dim

    • beta

      public double[] beta

    • zStats

      public double[] zStats

    • pValues

      public double[] pValues

    • numberOfIterations

      public int numberOfIterations

    • gradient

      protected final double[] gradient

    • hessian

      protected final double[][] hessian

    • delta

      protected final double[] delta

    • numberOfSamples

      protected int numberOfSamples

    • oneOver

      protected final double[] oneOver

    • gradMult

      protected final double[] gradMult

    • prod

      protected final double[] prod

    • sf

      protected final SigmoidFunction sf

  • Constructor Details

    • LogisticRegression

      protected LogisticRegression(int numberOfDepVars, int maximumNumberOfSamples)

  • Method Details

    • getGorLogisticRegressionObject

      public static LogisticRegression getGorLogisticRegressionObject(int numberOfDepVars, int maximumNumberOfSamples)

    • runRegression

      public boolean runRegression(double[][] x, boolean[] y, int numberOfSamples, double tol, int maxIter)

    • computeTestStats

      protected abstract void computeTestStats()

    • updateGradientAndHessian

      protected abstract void updateGradientAndHessian(double[][] x, boolean[] y)

    • updateGradientAndHessianFirstRun

      protected abstract void updateGradientAndHessianFirstRun(double[][] x, boolean[] y)