repicea.stats.integral

Class AdaptativeGaussHermiteQuadrature

java.lang.Object

repicea.stats.integral.NumericalIntegrationMethod

repicea.stats.integral.GaussQuadrature

repicea.stats.integral.GaussHermiteQuadrature

repicea.stats.integral.AdaptativeGaussHermiteQuadrature

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

LaplacianApproximation

public class AdaptativeGaussHermiteQuadrature
extends GaussHermiteQuadrature

The AdaptativeGaussHermiteQuadrature class implements the adaptative algorithm, i.e. the maximum of the function is first found before integrating.

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class repicea.stats.integral.GaussQuadrature

GaussQuadrature.NumberOfPoints

Field Summary

Fields inherited from class repicea.stats.integral.NumericalIntegrationMethod

rescalingFactors, weights, xValues

Constructor Summary

Constructors

Constructor and Description
AdaptativeGaussHermiteQuadrature(GaussQuadrature.NumberOfPoints numberOfPoints) Constructor.

Method Summary

Modifier and Type	Method and Description
double	getIntegralApproximation(AbstractMathematicalFunction functionToEvaluate, java.util.List<java.lang.Integer> parameterIndices, Matrix lowerCholeskyTriangle) This method returns the value of a multi-dimension integral
protected double	getOneDimensionIntegral(AbstractMathematicalFunction functionToEvaluate, java.lang.Integer parameterIndex, double standardDeviation) This method makes it possible to integrate an AbstractStatisticalExpression through Gauss-Hermite quadrature.

Methods inherited from class repicea.stats.integral.GaussHermiteQuadrature

getMultiDimensionIntegral, getRescalingFactors, getWeights, getXValues

Methods inherited from class repicea.stats.integral.GaussQuadrature

getOrderedNodes

Methods inherited from class repicea.stats.integral.NumericalIntegrationMethod

getLowerBound, getUpperBound, setLowerBound, setUpperBound, setXValuesFromListOfPoints

Methods inherited from class java.lang.Object

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

Constructor Detail

AdaptativeGaussHermiteQuadrature

public AdaptativeGaussHermiteQuadrature(GaussQuadrature.NumberOfPoints numberOfPoints)

Constructor.

Parameters:

numberOfPoints - the number of points the integral is based on.

Method Detail

getIntegralApproximation

public double getIntegralApproximation(AbstractMathematicalFunction functionToEvaluate,
                                       java.util.List<java.lang.Integer> parameterIndices,
                                       Matrix lowerCholeskyTriangle)

This method returns the value of a multi-dimension integral

Overrides:

getIntegralApproximation in class GaussHermiteQuadrature

Parameters:

functionToEvaluate - an EvaluableFunction instance that returns Double

parameterIndices - the indices of the parameters over which the integration is made

lowerCholeskyTriangle - the lower triangle of the Cholesky factorization of the variance-covariance matrix

Returns:

the approximation of the integral

getOneDimensionIntegral

protected double getOneDimensionIntegral(AbstractMathematicalFunction functionToEvaluate,
                                         java.lang.Integer parameterIndex,
                                         double standardDeviation)

Description copied from class: GaussHermiteQuadrature

This method makes it possible to integrate an AbstractStatisticalExpression through Gauss-Hermite quadrature.

Overrides:

getOneDimensionIntegral in class GaussHermiteQuadrature

Parameters:

functionToEvaluate - an EvaluableFunction instance that returns Double

parameterIndex - the index of the parameter over which the integration is made

standardDeviation - the standard deviation of this variable

Returns:

the approximation of the integral