Matrix-variate Distributions

Matrix-variate distributions are the distributions whose variate forms are Matrixvariate (i.e each sample is a matrix). Abstract types for matrix-variate distributions:

Common Interface

Both distributions implement the same set of methods:

size(d)

The size of each sample from the distribution d.

length(d)

The length (i.e number of elements) of each sample from the distribution d.

mean(d)

Return the mean matrix of d.

pdf(d, x)

Compute the probability density at the input matrix x.

logpdf(d, x)

Compute the logarithm of the probability density at the input matrix x.

rand(d)

Draw a sample matrix from the distribution d.

Wishart Distribution

The Wishart distribution is a multidimensional generalization of the Chi-square distribution, which is characterized by a degree of freedom ν, and a base matrix S.

Wishart(nu, S)    # Wishart distribution with nu degrees of freedom and base matrix S.

Inverse-Wishart Distribution

The Inverse Wishart distribution is usually used a the conjugate prior for the covariance matrix of a multivariate normal distribution, which is characterized by a degree of freedom ν, and a base matrix Φ.

InverseWishart(nu, P)    # Inverse-Wishart distribution with nu degrees of freedom and base matrix P.