com.rapidminer.tools.math
Class LinearRegression

java.lang.Object
  com.rapidminer.tools.math.LinearRegression

public class LinearRegression
extends java.lang.Object

This class can be used to calculate the coefficients of a (weighted) linear regression. It uses the class Matrix from the Jama package for this purpose. It is also possible to apply Ridge Regression which is a sort of regularization well suited especially for ill-posed problems. Please note that for the dependent matrix Y only one column is allowed.

Author:

Ingo Mierswa

Constructor Summary

LinearRegression()

Method Summary

static double[]	performRegression(Jama.Matrix x, Jama.Matrix y, double ridge) Calculates the coefficients of linear ridge regression.
static double[]	performRegression(Jama.Matrix x, Jama.Matrix y, double[] weights, double ridge) Performs a weighted linear ridge regression.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

LinearRegression

public LinearRegression()

Method Detail

performRegression

public static double[] performRegression(Jama.Matrix x,
                                         Jama.Matrix y,
                                         double[] weights,
                                         double ridge)

Performs a weighted linear ridge regression.

performRegression

public static double[] performRegression(Jama.Matrix x,
                                         Jama.Matrix y,
                                         double ridge)

Calculates the coefficients of linear ridge regression.