org.ojalgo.random.process

Class GaussianProcess

java.lang.Object

org.ojalgo.random.process.GaussianProcess

All Implemented Interfaces:

RandomProcess<Normal>

public final class GaussianProcess
extends Object

A Gaussian process is a stochastic process whose realizations consist of random values associated with every point in a range of times (or of space) such that each such random variable has a normal distribution. Moreover, every finite collection of those random variables has a multivariate normal distribution. Prior to calling getDistribution(double) or simulate(int, int, double) you must call addObservation(Double, double) one or more times.

Author:

apete

Nested Class Summary

Nested classes/interfaces inherited from interface org.ojalgo.random.process.RandomProcess

RandomProcess.SimulationResults

Constructor Summary

Constructors

Constructor and Description
GaussianProcess(GaussianField.Covariance<Double> covarFunc)
GaussianProcess(GaussianField.Mean<Double> meanFunc, GaussianField.Covariance<Double> covarFunc)

Method Summary

Modifier and Type	Method and Description
boolean	addObservation(Double x, double y)
void	calibrate()
Normal1D	getDistribution(Double... evaluationPoint)
Normal	getDistribution(double evaluationPoint)
double	getExpected() Equivalent to calling RandomProcess.getDistribution(double) with argumant 1.0, and then Distribution.getExpected().
double	getLowerConfidenceQuantile(double confidence) The same thing can be achieved by first calling RandomProcess.getDistribution(double) with argumant 1.0, and then ContinuousDistribution.getQuantile(double) (but with different input argument).
protected double	getNormalisedRandomIncrement()
double	getStandardDeviation() Equivalent to calling RandomProcess.getDistribution(double) with argumant 1.0, and then Distribution.getStandardDeviation().
double	getUpperConfidenceQuantile(double confidence) The same thing can be achieved by first calling RandomProcess.getDistribution(double) with argumant 1.0, and then ContinuousDistribution.getQuantile(double) (but with different input argument).
double	getValue()
double	getVariance() Equivalent to calling RandomProcess.getDistribution(double) with argumant 1.0, and then Distribution.getVariance().
protected void	setObservations(Collection<? extends ComparableToDouble<Double>> c)
void	setValue(double newValue)
RandomProcess.SimulationResults	simulate(int numberOfRealisations, int numberOfSteps, double stepSize)
protected double	step(double currentValue, double stepSize, double normalisedRandomIncrement)

Methods inherited from class java.lang.Object

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

Constructor Detail

GaussianProcess

public GaussianProcess(GaussianField.Covariance<Double> covarFunc)

GaussianProcess

public GaussianProcess(GaussianField.Mean<Double> meanFunc,
                       GaussianField.Covariance<Double> covarFunc)

Method Detail

calibrate

public void calibrate()

getDistribution

public Normal getDistribution(double evaluationPoint)

Parameters:

evaluationPoint - How far into the future?

Returns:

The distribution for the process value at that future time.

getDistribution

public Normal1D getDistribution(Double... evaluationPoint)

getNormalisedRandomIncrement

protected double getNormalisedRandomIncrement()

step

protected double step(double currentValue,
                      double stepSize,
                      double normalisedRandomIncrement)

addObservation

public final boolean addObservation(Double x,
                                    double y)

getExpected

public final double getExpected()

Equivalent to calling RandomProcess.getDistribution(double) with argumant 1.0, and then Distribution.getExpected().

getLowerConfidenceQuantile

public final double getLowerConfidenceQuantile(double confidence)

The same thing can be achieved by first calling RandomProcess.getDistribution(double) with argumant 1.0, and then ContinuousDistribution.getQuantile(double) (but with different input argument).

getStandardDeviation

public final double getStandardDeviation()

Equivalent to calling RandomProcess.getDistribution(double) with argumant 1.0, and then Distribution.getStandardDeviation().

getUpperConfidenceQuantile

public final double getUpperConfidenceQuantile(double confidence)

The same thing can be achieved by first calling RandomProcess.getDistribution(double) with argumant 1.0, and then ContinuousDistribution.getQuantile(double) (but with different input argument).

getValue

public final double getValue()

getVariance

public final double getVariance()

Equivalent to calling RandomProcess.getDistribution(double) with argumant 1.0, and then Distribution.getVariance().

setValue

public final void setValue(double newValue)

simulate

public final RandomProcess.SimulationResults simulate(int numberOfRealisations,
                                                      int numberOfSteps,
                                                      double stepSize)

Specified by:

simulate in interface RandomProcess<D extends Distribution>

Returns:

An array of sample sets. The array has aNumberOfSteps elements, and each sample set has aNumberOfRealisations samples.

setObservations

protected final void setObservations(Collection<? extends ComparableToDouble<Double>> c)