The Analysis of Variance for the Two-Way Classification Fixed-Effects Model with Observations Within a Row Serially Correlated
行内观测序列相关的双向固定效应方差分析
This paper considers a Bayesian approach to the two-way analysis of variance when the error terms within rows are not independent, but have a covariance matrix dependent on at most two unknown parameters. The posterior density for a linear contrast of row means is obtained. The robustness of the procedure to different priors on the serial correlation coefficient is also discussed.