Answers are grounded in this paper and cite their sources.
We consider some estimation and distribution problems encountered in a two way analysis of variance model with only one observation per cell, errors correlated in one level, and the variances are not necessarily equal. The independence criteria for the row and interaction mean sum of squares and distribution of the maximum likelihood estimator of the correlation coefficient are given.
No takes yet. Share an insight, caveat, or question.
Ngwengwe et al. (1991) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: