rmatnorm: Matrix normal distribution

In jijiadong/MVDN: Matrix-variate Differential Network Analysis

Description

The function provide a random number generator for the Matrix normal distribution vec(X) ~ N(vec(M), kron(sigmaT, sigmaS)).

Usage

rmatnorm(n, mean = matrix(0, nrow = nrow(sigmaS), ncol = ncol(sigmaT)), sigmaS, sigmaT, method = c("eigen", "svd", "chol"))

Arguments

n	number of observations.
mean	mean matrix, default is matrix(0, nrow=nrow(sigmaS), ncol=ncol(sigmaT)).
sigmaS	covariance matrix between rows.
sigmaT	covariance matrix between columns.
method	string specifying the matrix decomposition used todetermine the matrix root of sigmaS and sigmaT. Possible methods are eigenvalue decomposition ("eigen", default), singular value decomposition ("svd"), and Cholesky decomposition ("chol").

Value

returns a p*q*n array if n > 1. returns a p*q matrix if n = 1.

See Also

rmatdata

Examples

p <- 5; q <- 6; n <- 10
set.seed(1212)
s1 <- 0.5^abs(outer(1:p,1:p,"-"))
s2 <- 0.4^abs(outer(1:q,1:q,"-"))
x <- rmatnorm(n,mean=matrix(1,p,q),sigmaS=s1,sigmaT=s2)

jijiadong/MVDN documentation built on Sept. 6, 2020, 7:15 p.m.