gov.sandia.cognition.statistics.montecarlo

Interface MonteCarloIntegrator<OutputType>

Type Parameters:

OutputType - Type of output from the integrator.

All Superinterfaces:

java.lang.Cloneable, CloneableSerializable, java.io.Serializable

All Known Implementing Classes:

MultivariateMonteCarloIntegrator, UnivariateMonteCarloIntegrator

@PublicationReference(author={"Christian P. Robert","George Casella"},title="Monte Carlo Statistical Methods, Second Edition",type=Book,year=2004,pages={83,106}) @PublicationReference(author="Wikipedia",title="Monte Carlo integration",type=WebPage,year=2010,url="http://en.wikipedia.org/wiki/Monte_Carlo_integration")
public interface MonteCarloIntegrator<OutputType>
extends CloneableSerializable

Monte Carlo integration is a way of compute the integral of a function using samples from another. If, as is typical, the samples come from a probability distribution, then the result of Monte Carlo integration is the expectation of the function under the probability distribution. That is, Expectation == integral( g(x) * p(x) dx ) ~= (1/n) sum(g(x)), where x ~ p(x).

Since:

3.0

Author:

Kevin R. Dixon

Method Summary

Modifier and Type	Method and Description
Distribution<? extends OutputType>	getMean(java.util.Collection<? extends OutputType> samples) Computes the Monte Carlo distribution of the given samples.
Distribution<? extends OutputType>	getMean(java.util.List<? extends WeightedValue<? extends OutputType>> samples) Computes the Monte Carlo distribution of the given weighted samples.
<SampleType> Distribution<? extends OutputType>	integrate(java.util.Collection<? extends SampleType> samples, Evaluator<? super SampleType,? extends OutputType> expectationFunction) Integrates the given function given samples from another function.
<SampleType> Distribution<? extends OutputType>	integrate(java.util.List<? extends WeightedValue<? extends SampleType>> samples, Evaluator<? super SampleType,? extends OutputType> expectationFunction) Integrates the given function given weighted samples from another function.

Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable

clone

Method Detail

integrate

<SampleType> Distribution<? extends OutputType> integrate(java.util.Collection<? extends SampleType> samples,
                                                          Evaluator<? super SampleType,? extends OutputType> expectationFunction)

Integrates the given function given samples from another function.

Type Parameters:

SampleType - Type of samples to consider.

Parameters:

samples - Samples from the underlying distribution.

expectationFunction - Function for which to compute the expectation.

Returns:

Distribution of the integration.

integrate

<SampleType> Distribution<? extends OutputType> integrate(java.util.List<? extends WeightedValue<? extends SampleType>> samples,
                                                          Evaluator<? super SampleType,? extends OutputType> expectationFunction)

Integrates the given function given weighted samples from another function.

Type Parameters:

SampleType - Type of samples to consider.

Parameters:

samples - Weighted samples from the underlying distribution.

expectationFunction - Function for which to compute the expectation.

Returns:

Distribution of the integration.

getMean

Distribution<? extends OutputType> getMean(java.util.Collection<? extends OutputType> samples)

Computes the Monte Carlo distribution of the given samples.

Parameters:

samples - Samples to consider.

Returns:

Distribution describing the samples.

getMean

Distribution<? extends OutputType> getMean(java.util.List<? extends WeightedValue<? extends OutputType>> samples)

Computes the Monte Carlo distribution of the given weighted samples.

Parameters:

samples - Weighted samples to consider.

Returns:

Distribution describing the samples.