Viana M A
Department of Ophthalmology and Visual Sciences, University of Illinois at Chicago 60612-7243.
Biometrics. 1994 Sep;50(3):813-20.
Combined maximum likelihood estimates for equicorrelation covariance matrices are considered. The case of a common equicorrelation rho and possibly different standard deviations sigma 1, ..., sigma k among k experimental groups is examined first, and the estimation of (rho, sigma 1, ..., sigma k) is discussed. Second, under the assumption of a common standard deviation and possibly different equicorrelations, the estimation of (rho 1, ..., rho k, sigma) is considered. In each case, maximum likelihood solutions and corresponding large-sample variances are presented.