Maximum likelihood estimation in a two-way analysis of variance with correlated errors in one classification

Abstract

A two-way analysis of variance model with correlated errors in one classification is discussed. It is assumed that the p measurements in each row have a general covariance matrix Σ The maximum likelihood estimates of the row and column parameters as well as of the covariance matrix are obtained and their asymptotic distribution are discussed. Finally, the likelihood ratio test statistics for the various hypotheses in a two-way layout are given.