Answers are grounded in this paper and cite their sources.
SUMMARY Linear models used for estimation of variance components are generally formulated under the assumption of independent errors. Herein we consider a two-way (mixed) model with errors assumed to follow a first-order, nonstationary autoregressive process. Maximum likelihood techniques are employed to estimate the variance components and also a parameter for the autoregression. An example using lactational yield of cows is given along with some Monte Carlo results.
No takes yet. Share an insight, caveat, or question.
Mansour et al. (1985) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: