weka.core.matrix

Class LinearRegression

java.lang.Object

weka.core.matrix.LinearRegression

All Implemented Interfaces:

RevisionHandler

public class LinearRegression
extends Object
implements RevisionHandler

Class for performing (ridged) linear regression.

Version:

$Revision: 1.4 $

Author:

Fracpete (fracpete at waikato dot ac dot nz)

Field Summary

Fields

Modifier and Type	Field and Description
protected double[]	m_Coefficients the coefficients

Constructor Summary

Constructors

Constructor and Description
LinearRegression(Matrix a, Matrix y, double ridge) Performs a (ridged) linear regression.
LinearRegression(Matrix a, Matrix y, double[] w, double ridge) Performs a weighted (ridged) linear regression.

Method Summary

Modifier and Type	Method and Description
protected void	calculate(Matrix a, Matrix y, double ridge) performs the actual regression.
double[]	getCoefficients() returns the calculated coefficients
String	getRevision() Returns the revision string.
String	toString() returns the coefficients in a string representation

Methods inherited from class java.lang.Object

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

Field Detail

m_Coefficients

protected double[] m_Coefficients

the coefficients

Constructor Detail

LinearRegression

public LinearRegression(Matrix a,
                        Matrix y,
                        double ridge)

Performs a (ridged) linear regression.

Parameters:

a - the matrix to perform the regression on

y - the dependent variable vector

ridge - the ridge parameter

Throws:

IllegalArgumentException - if not successful

LinearRegression

public LinearRegression(Matrix a,
                        Matrix y,
                        double[] w,
                        double ridge)

Performs a weighted (ridged) linear regression.

Parameters:

a - the matrix to perform the regression on

y - the dependent variable vector

w - the array of data point weights

ridge - the ridge parameter

Throws:

IllegalArgumentException - if the wrong number of weights were provided.

Method Detail

calculate

protected void calculate(Matrix a,
                         Matrix y,
                         double ridge)

performs the actual regression.

Parameters:

a - the matrix to perform the regression on

y - the dependent variable vector

ridge - the ridge parameter

Throws:

IllegalArgumentException - if not successful

getCoefficients

public final double[] getCoefficients()

returns the calculated coefficients

Returns:

the coefficients

toString

public String toString()

returns the coefficients in a string representation

Overrides:

toString in class Object

getRevision

public String getRevision()

Returns the revision string.

Specified by:

getRevision in interface RevisionHandler

Returns:

the revision