Title: Bayesian minimax estimation for means in $k$ sample problems
Authors: Ryo Imai - University of Tokyo (Japan) [presenting]
Abstract: The simultaneous estimation of means of $k$ multivariate normal populations is considered when one suspects that the $k$ means are nearly equal. As an alternative to the preliminary test estimator based on the test statistics for testing hypothesis of equal means, we derive Bayesian and minimax estimators which shrink individual sample means toward a pooled mean estimator given under the hypothesis. It is shown that both the preliminary test estimator and the Bayesian minimax shrinkage estimators are further improved by shrinking the pooled mean estimator. The performance of the proposed shrinkage estimators is investigated by simulation. We will also discuss individual estimation of a mean of one sample and Efron-Morris type estimation of a mean matrix in the matrix normal model.