fr.cnes.sirius.patrius.math.stat.regression

Class GLSMultipleLinearRegression

java.lang.Object

fr.cnes.sirius.patrius.math.stat.regression.AbstractMultipleLinearRegression

fr.cnes.sirius.patrius.math.stat.regression.GLSMultipleLinearRegression

All Implemented Interfaces:

MultipleLinearRegression

public class GLSMultipleLinearRegression
extends AbstractMultipleLinearRegression

The GLS implementation of the multiple linear regression. GLS assumes a general covariance matrix Omega of the error

 u ~ N(0, Omega)
 

Estimated by GLS,

 b=(X' Omega^-1 X)^-1X'Omega^-1 y
 

whose variance is

 Var(b)=(X' Omega^-1 X)^-1
 

Since:

2.0

Version:

$Id: GLSMultipleLinearRegression.java 18108 2017-10-04 06:45:27Z bignon $

Constructor Summary

Constructors

Constructor and Description
GLSMultipleLinearRegression()

Method Summary

Modifier and Type	Method and Description
protected RealVector	calculateBeta() Calculates beta by GLS.
protected RealMatrix	calculateBetaVariance() Calculates the variance on the beta.
protected double	calculateErrorVariance() Calculates the estimated variance of the error term using the formula Var(u) = Tr(u' Omega^-1 u)/(n-k) where n and k are the row and column dimensions of the design matrix X.
protected RealMatrix	getOmegaInverse() Get the inverse of the covariance.
protected void	newCovarianceData(double[][] omegaIn) Add the covariance data.
void	newSampleData(double[] y, double[][] x, double[][] covariance) Replace sample data, overriding any previous sample.

Methods inherited from class fr.cnes.sirius.patrius.math.stat.regression.AbstractMultipleLinearRegression

calculateResiduals, calculateYVariance, estimateErrorVariance, estimateRegressandVariance, estimateRegressionParameters, estimateRegressionParametersStandardErrors, estimateRegressionParametersVariance, estimateRegressionStandardError, estimateResiduals, getX, getY, isNoIntercept, newSampleData, newXSampleData, newYSampleData, setNoIntercept, validateCovarianceData, validateSampleData

Methods inherited from class java.lang.Object

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

Constructor Detail

GLSMultipleLinearRegression

public GLSMultipleLinearRegression()

Method Detail

newSampleData

public void newSampleData(double[] y,
                          double[][] x,
                          double[][] covariance)

Replace sample data, overriding any previous sample.

Parameters:

y - y values of the sample

x - x values of the sample

covariance - array representing the covariance matrix

newCovarianceData

protected void newCovarianceData(double[][] omegaIn)

Add the covariance data.

Parameters:

omegaIn - the [n,n] array representing the covariance

getOmegaInverse

protected RealMatrix getOmegaInverse()

Get the inverse of the covariance.

The inverse of the covariance matrix is lazily evaluated and cached.

Returns:

inverse of the covariance

calculateBeta

protected RealVector calculateBeta()

Calculates beta by GLS.

  b=(X' Omega^-1 X)^-1X'Omega^-1 y
 

Specified by:

calculateBeta in class AbstractMultipleLinearRegression

Returns:

beta

calculateBetaVariance

protected RealMatrix calculateBetaVariance()

Calculates the variance on the beta.

  Var(b)=(X' Omega^-1 X)^-1
 

Specified by:

calculateBetaVariance in class AbstractMultipleLinearRegression

Returns:

The beta variance matrix

calculateErrorVariance

protected double calculateErrorVariance()

Calculates the estimated variance of the error term using the formula

  Var(u) = Tr(u' Omega^-1 u)/(n-k)
 

where n and k are the row and column dimensions of the design matrix X.

Overrides:

calculateErrorVariance in class AbstractMultipleLinearRegression

Returns:

error variance

Since:

2.2