com.itemanalysis.psychometrics.analysis

Class AbstractMultivariateFunction

java.lang.Object

com.itemanalysis.psychometrics.analysis.AbstractMultivariateFunction

All Implemented Interfaces:

org.apache.commons.math3.analysis.MultivariateFunction

Direct Known Subclasses:

ExampleMultivariateFunction, HaebaraMethod, StockingLordMethod

public abstract class AbstractMultivariateFunction
extends java.lang.Object
implements org.apache.commons.math3.analysis.MultivariateFunction

Abstract class that includes methods for numeric computations of teh gradientAt, partial derivatives, and hessian. Extend this class with an implementation of the rho(double[] x) method as specified in the MultivariateRealFunction interface.

Constructor Summary

Constructors

Constructor and Description
AbstractMultivariateFunction()
AbstractMultivariateFunction(double EPSILON)

Method Summary

Modifier and Type	Method and Description
org.apache.commons.math3.analysis.MultivariateVectorFunction	gradient()
double[]	gradientAt(double[] x) Numerically compute gradientAt by the central difference method.
org.apache.commons.math3.linear.RealMatrix	hessianAt(double[] x) Numerically compute Hessian using a finite difference method.
double	machineEpsilon()
org.apache.commons.math3.analysis.MultivariateFunction	partialDerivative(int k)

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.commons.math3.analysis.MultivariateFunction

value

Constructor Detail

AbstractMultivariateFunction

public AbstractMultivariateFunction()

AbstractMultivariateFunction

public AbstractMultivariateFunction(double EPSILON)

Method Detail

machineEpsilon

public double machineEpsilon()

gradient

public org.apache.commons.math3.analysis.MultivariateVectorFunction gradient()

partialDerivative

public org.apache.commons.math3.analysis.MultivariateFunction partialDerivative(int k)

gradientAt

public double[] gradientAt(double[] x)

Numerically compute gradientAt by the central difference method. Override this method when the analytic gradientAt is available.

Parameters:

x -

Returns:

hessianAt

public org.apache.commons.math3.linear.RealMatrix hessianAt(double[] x)

Numerically compute Hessian using a finite difference method. Override this method when the analytic Hessian is available.

Parameters:

x -

Returns: