Package org.apache.commons.rng.sampling.distribution

Class TSampler

java.lang.Object

org.apache.commons.rng.sampling.distribution.TSampler

All Implemented Interfaces:

ContinuousSampler, SharedStateContinuousSampler, SharedStateSampler<SharedStateContinuousSampler>

Direct Known Subclasses:

TSampler.NormalTSampler, TSampler.StudentsTSampler

public abstract class TSampler
extends java.lang.Object
implements SharedStateContinuousSampler

Sampling from a T distribution.

Uses Bailey's algorithm for t-distribution sampling:

 Bailey, R. W. (1994)
 "Polar Generation of Random Variates with the t-Distribution."
 Mathematics of Computation 62, 779-781.
 

Sampling uses UniformRandomProvider.nextLong().

Since:

1.5

See Also:

Student's T distribution (wikipedia), Mathematics of Computation, 62, 779-781

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
private static class	TSampler.NormalTSampler	Sample from a t-distribution using a normal distribution.
private static class	TSampler.StudentsTSampler	Sample from a t-distribution using Bailey's algorithm.

Field Summary

Fields

Modifier and Type	Field	Description
private static double	HUGE_DF	Threshold for huge degrees of freedom.
private UniformRandomProvider	rng	Source of randomness.

Constructor Summary

Constructors

Constructor	Description
TSampler(UniformRandomProvider rng)

Method Summary

Modifier and Type	Method	Description
(package private) long	nextLong()	Generates a long value.
static TSampler	of(UniformRandomProvider rng, double degreesOfFreedom)	Create a new t distribution sampler.
java.lang.String	toString()
abstract TSampler	withUniformRandomProvider(UniformRandomProvider rng)	Create a new instance of the sampler with the same underlying state using the given uniform random provider as the source of randomness.

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.commons.rng.sampling.distribution.ContinuousSampler

sample, samples, samples

Field Detail

HUGE_DF

private static final double HUGE_DF

Threshold for huge degrees of freedom. Above this value the CDF of the t distribution matches the normal distribution. Value is 2/eps (where eps is the machine epsilon) or approximately 9.0e15.

See Also:

Constant Field Values

rng

private final UniformRandomProvider rng

Source of randomness.

Constructor Detail

TSampler

TSampler(UniformRandomProvider rng)

Parameters:

rng - Generator of uniformly distributed random numbers.

Method Detail

withUniformRandomProvider

public abstract TSampler withUniformRandomProvider(UniformRandomProvider rng)

Create a new instance of the sampler with the same underlying state using the given uniform random provider as the source of randomness.

Specified by:

withUniformRandomProvider in interface SharedStateSampler<SharedStateContinuousSampler>

Parameters:

rng - Generator of uniformly distributed random numbers.

Returns:

the sampler

nextLong

long nextLong()

Generates a long value. Used by algorithm implementations without exposing access to the RNG.

Returns:

the next random value

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

of

public static TSampler of(UniformRandomProvider rng,
                          double degreesOfFreedom)

Create a new t distribution sampler.

Parameters:

rng - Generator of uniformly distributed random numbers.

degreesOfFreedom - Degrees of freedom.

Returns:

the sampler

Throws:

java.lang.IllegalArgumentException - if degreesOfFreedom <= 0