gov.sandia.cognition.statistics.bayesian

Class BayesianCredibleInterval

java.lang.Object

gov.sandia.cognition.util.AbstractCloneableSerializable

gov.sandia.cognition.statistics.method.ConfidenceInterval

gov.sandia.cognition.statistics.bayesian.BayesianCredibleInterval

All Implemented Interfaces:

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

@PublicationReference(author="Wikipedia",
                      title="Credible interval",
                      type=WebPage,
                      year=2010,
                      url="http://en.wikipedia.org/wiki/Credible_interval")
public class BayesianCredibleInterval
extends ConfidenceInterval

A Bayesian credible interval defines a bound that a scalar parameter is within the given interval. This is the Bayesian analogue to the frequentist notion of a confidence interval.

Since:

3.0

Author:

Kevin R. Dixon

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
BayesianCredibleInterval(double median, double lowerBound, double upperBound, double confidence) Creates a new instance of ConfidenceInterval

Method Summary

Modifier and Type	Method and Description
static <NumberType extends java.lang.Number> BayesianCredibleInterval	compute(UnivariateDistribution<NumberType> distribution, double confidence) Creates a Bayesian credible interval by inverting the given CDF.

Methods inherited from class gov.sandia.cognition.statistics.method.ConfidenceInterval

getCentralValue, getConfidence, getLowerBound, getNumSamples, getUpperBound, setCentralValue, setConfidence, setLowerBound, setNumSamples, setUpperBound, toString, withinInterval

Methods inherited from class gov.sandia.cognition.util.AbstractCloneableSerializable

clone

Methods inherited from class java.lang.Object

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

Constructor Detail

BayesianCredibleInterval

public BayesianCredibleInterval(double median,
                                double lowerBound,
                                double upperBound,
                                double confidence)

Creates a new instance of ConfidenceInterval

Parameters:

median - Median value of the parameter

lowerBound - Lower bound for the parameter

upperBound - Upper bound for the parameter

confidence - Confidence that the parameter is within the bound, or 1-pvalue, on the interval [0,1], where confidence=0 means the bound will be zero, while confidence=1 means that the confidence bound will be the entire support of the underlying distribution.

Method Detail

compute

public static <NumberType extends java.lang.Number> BayesianCredibleInterval compute(UnivariateDistribution<NumberType> distribution,
                                                                                     double confidence)

Creates a Bayesian credible interval by inverting the given CDF.

Type Parameters:

NumberType - Type of number to consider

Parameters:

distribution - Distribution to compute the credible interval of, typically the posterior from a Bayesian estimation algorithm.

confidence - Confidence that the parameter is within the bound, or 1-pvalue, on the interval [0,1], where confidence=0 means the bound will be zero, while confidence=1 means that the confidence bound will be the entire support of the underlying distribution.

Returns:

Confidence interval of the parameter given the distribution