Dq. Wang et Cro. Lawoko, ESTIMATION OF PARAMETERS FOR NORMALLY DISTRIBUTED RANDOM MATRICES, Linear algebra and its applications, 210, 1994, pp. 199-208
Consider the p x n random matrix X which is normally distributed with
mean M, and let the covariance matrix between any two columns of X (sa
y x(i) and x(j)) be gamma(ij)(Sigma + Sigma(epsilon)). Expectations an
d covariances of the maximum likelihood estimators of M and Sigma are
given for the situation when gamma(ij) and Sigma(epsilon) are known.