On maximum likelihood estimators in certain multivariate

Rok, strany: 1999, 111 - 119

We consider the problem of maximum likelihood (ML) estimation of variances in certain mixed linear models with two variance components. For those models a procedure to obtain ML estimators is described. The solutions produced by this procedure are described. It is shown that under some conditions the solutions of the ML equations cannot give ML estimators i.e., estimators which maximize the likelihood function. For those models necessary and sufficient conditions are given, for which the solutions of the ML equations lead to ML estimators. Two way layouts model and equicorrelated model are considered in details.

ISO 690:

Gnot, S., Stemann, D., Trenkler, G., Urbańska-Motyka, A. 1999. On maximum likelihood estimators in certain multivariate. In Tatra Mountains Mathematical Publications, vol. 17, no.3, pp. 111-119. 1210-3195.

APA:

Gnot, S., Stemann, D., Trenkler, G., Urbańska-Motyka, A. (1999). On maximum likelihood estimators in certain multivariate. Tatra Mountains Mathematical Publications, 17(3), 111-119. 1210-3195.