com.numericalmethod.suanshu.stats.random.multivariate
Class Gaussian

java.lang.Object
  com.numericalmethod.suanshu.stats.random.multivariate.Gaussian

All Implemented Interfaces:

RandomVectorGenerator

public class Gaussian
extends java.lang.Object
implements RandomVectorGenerator

This class generates multivariate Normal random numbers by implementing the "Best Approximation of Indefinite Matrices" in the reference.

In the case where the covariance matrix is not positive definite, we force the diagonal elements in the eigen decomposition to a small non-negative number, e.g., 0.

Then, we re-construct a positive definite matrix from the new diagonal elements.

The R equivalent function is rmvnorm in package mvtnorm.

See Also:

• Wikipedia: Multivariate normal distribution
• "Jin Wang, Chunlei Liu. "Generating Multivariate Mixture of Normal Distributions using a Modified Cholesky Decomposition," Simulation Conference, 2006. WSC 06. Proceedings of the Winter. p. 342 - 347. 3-6 Dec. 2006."

Constructor Summary

Gaussian(Vector mu, Matrix sigma)
Construct a multivariate Gaussian random vector generator.

Gaussian(Vector mu, Matrix sigma, RandomNumberGenerator rng, double epsilon)
Construct a multivariate Gaussian random vector generator.

Method Summary

double[]	nextVector() Get the next random vector.
void	seed(long... seeds) Seed the random number generator to produce repeatable sequences.

Methods inherited from class java.lang.Object

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

Constructor Detail

Gaussian

public Gaussian(Vector mu,
                Matrix sigma,
                RandomNumberGenerator rng,
                double epsilon)

Construct a multivariate Gaussian random vector generator.

Parameters:

mu - the mean

sigma - the covariance matrix

rng - a uniform random number generator

epsilon - a precision parameter: when a number |x| ≤ ε, it is considered 0

Gaussian

public Gaussian(Vector mu,
                Matrix sigma)

Construct a multivariate Gaussian random vector generator.

Parameters:

mu - the mean

sigma - the covariance matrix

Method Detail

seed

public void seed(long... seeds)

Description copied from interface: RandomVectorGenerator

Seed the random number generator to produce repeatable sequences.

Specified by:

seed in interface RandomVectorGenerator

Parameters:

seeds - the seeds

nextVector

public double[] nextVector()

Description copied from interface: RandomVectorGenerator

Get the next random vector.

Specified by:

nextVector in interface RandomVectorGenerator

Returns:

the next random vector