STATISTICA Visual Basic Library of Matrix Functions - Wishart

Method:

Environment:

STATISTICA Visual Basic (SVB)

Function:

Generates a random sample of size n drawn from a p-dimensional multivariate normal distribution, with arbitrary mean and full covariance matrix.

Input arguments:

n � Number of samples to be drawn from the multivariate normal distribution.

� A vector of p means for the multivariate normal variables.

� The p x p covariance matrix for the multivariate normal distribution. This matrix must be positive definite.

seed � An optional seed parameter for the random number generator. Must be a positive integer. If not supplied the function will default to the  STATISTICA system random number seed and sequence. Thus, the random number generator can be seeded programmatically from within STATISTICA Visual Basic (SVB), for example, to create a replicable sequence of random samples (call the SVB function Randomize([ByVal Seed] before generating sequences of samples using the MatrixDrawMVN function without the optional seed argument)

Output:

Returns an n x p matrix S storing a random sample of size n drawn from a p-dimensional multivariate normal distribution with mean and covariance matrix .

Usage:

The function can be called with the proper input arguments as above. If the last argument seed was not supplied the system clock will be used as seed.

 

Method:

Environment:

STATISTICA Visual Basic (SVB)

Function:

Generates a random sample drawn from an inverse Wishart distribution with dimension p, df degrees of freedom and covariance matrix .

Input arguments:

df � The degrees of freedom for the inverse Wishart distribution from which the random sample is drawn.

� The covariance matrix of multivariate normal variables.

seed � An optional argument for the seed for the random number generator. Given a constant seed, repeated calls to the function will generate identical samples. If not supplied, the STATISTICA system random number seed and sequence will be applied. Thus, the random number generator can be seeded programmatically from within STATISTICA Visual Basic (SVB), for example, to create a replicable sequence of random draws (call the SVB function Randomize([ByVal Seed] before generating sequences of random draws using the MatrixDrawInverseWishart function without the optional seed argument)

Output:

Returns a matrix storing a random sample  drawn from an inverse Wishart sample

Usage:

This function draws a random sample from an inverse Wishart distribution with . The returned sample can also be converted to a random sample S, with , drawn from a Wishart distribution via the relation .