Regular Article
Other Classes of Minimax Estimators of Variance Covariance Matrix in Multivariate Normal Distribution

https://doi.org/10.1006/jmva.2000.1934Get rights and content
Under an Elsevier user license
open archive

Abstract

It is well known that the best equivariant estimator of the variance covariance matrix of the multivariate normal distribution with respect to the full affine group of transformation is not even minimax. Some minimax estimators have been proposed. Here we treat this problem in the framework of a multivariate analysis of variance (MANOVA) model and give other classes of minimax estimators.

Keywords

conditional risk difference, entropy loss, inadmissibility, MANOVA, minimax estimation, Wishart matrix

Cited by (0)

1

Correspondence should be addressed to Department of Geosystem Engineering, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan. E-mail: hara@geosys. t.u-tokyo.ac.jp.