repicea.stats.distributions

Class StudentTDistribution

java.lang.Object

repicea.stats.distributions.StudentTDistribution

All Implemented Interfaces:

java.io.Serializable, CentralMomentsGettable, Distribution

public class StudentTDistribution
extends java.lang.Object
implements Distribution

This class implements the standard Student's t distribution. The algorithm behind the random deviate generation is that of Bailey (1994) based on polar generation.

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from interface repicea.stats.Distribution

Distribution.Type

Constructor Summary

Constructors

Constructor and Description
StudentTDistribution(int degreesOfFreedom) Constructor for a standard Student's distribution, i.e.
StudentTDistribution(Matrix mu, Matrix variance, int degreesOfFreedom) This constructor creates a Student't distribution with mean mu and variance sigma2.

Method Summary

Modifier and Type	Method and Description
Matrix	getMean() This method returns the first central moment, i.e.
Matrix	getRandomRealization() This method draws a random realization from the distribution.
Matrix	getStandardDeviation() This method returns the lower triangle of the Cholesky decomposition of the variance-covariance matrix.
Distribution.Type	getType() This method returns the type of the distribution.
Matrix	getVariance() This method returns the second central moment, i.e.
boolean	isMultivariate() This method returns true if the GaussianFunction instance is multivariate.
boolean	isParametric() This method returns true if the distribution is parametric or false otherwise.

Methods inherited from class java.lang.Object

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

Methods inherited from interface repicea.stats.Distribution

isUnivariate

Constructor Detail

StudentTDistribution

public StudentTDistribution(Matrix mu,
                            Matrix variance,
                            int degreesOfFreedom)

This constructor creates a Student't distribution with mean mu and variance sigma2. NOTE: It does not handle multivariate distribution. If the mean and the variance are of dimensions > 1, they are assumed to be independent, i.e. the non zero off diagonal elements in the variance matrix will not be considered.

Parameters:

mu - the mean of the function

variance - the variance of the distribution

degreesOfFreedom - the degrees of freedom of the distribution

StudentTDistribution

public StudentTDistribution(int degreesOfFreedom)

Constructor for a standard Student's distribution, i.e. with mean 0 and sigma^2 = 1.

Parameters:

degreesOfFreedom -

Method Detail

isMultivariate

public boolean isMultivariate()

Description copied from interface: Distribution

This method returns true if the GaussianFunction instance is multivariate.

Specified by:

isMultivariate in interface Distribution

Returns:

a boolean

getRandomRealization

public Matrix getRandomRealization()

Description copied from interface: Distribution

This method draws a random realization from the distribution.

Specified by:

getRandomRealization in interface Distribution

Returns:

the observation in a Matrix instance

getStandardDeviation

public Matrix getStandardDeviation()

This method returns the lower triangle of the Cholesky decomposition of the variance-covariance matrix.

Returns:

a Matrix instance

getMean

public Matrix getMean()

Description copied from interface: CentralMomentsGettable

This method returns the first central moment, i.e. the mean vector, of the random variable.

Specified by:

getMean in interface CentralMomentsGettable

Returns:

a Matrix instance

getVariance

public Matrix getVariance()

Description copied from interface: CentralMomentsGettable

This method returns the second central moment, i.e. the variance-covariance matrix, of the random variable.

Specified by:

getVariance in interface CentralMomentsGettable

Returns:

a Number instance

getType

public Distribution.Type getType()

Description copied from interface: Distribution

This method returns the type of the distribution.

Specified by:

getType in interface Distribution

Returns:

a Type enum

isParametric

public boolean isParametric()

Description copied from interface: Distribution

This method returns true if the distribution is parametric or false otherwise.

Specified by:

isParametric in interface Distribution

Returns:

a boolean