Inadmissibility of the usual estimator for the location parameters of spherically symmetric distributions
It is well-known that the usual estimator, the sample mean, for the mean of a multivariate normal distribution is inadmissible. Stein's ideas and results have been generalized and extended in two main directions, namely, to more general loss functions, and to other distributions with location parameters. Among these extensions are \textit{A. R. Cohen Brandwein} and \textit{W. E. Strawderman}'s papers [Ann. of Statist. 6, 377- 416 (1978; Zbl 0402.62019) and ibid. 8, 279-284 (1980; Zbl 0432.62008)], in which the inadmissibility of the usual estimator for the location parameters of spherically symmetric distributions and other extensions are proposed. It is desirable to shorten the proofs given by Brandwein and Strawderman, and this report is designed to achieve this goal. Besides, some minimax results are also improved.
- Estimation with quadratic loss.
- Admissible Bayes equivariant estimation of location vectors for spherically symmetric distributions with unknown scale
- Estimation of location parameters for spherically symmetric distributions
- scientific article; zbMATH DE number 1763556
- Admissibility and minimaxity of generalized Bayes estimators for spherically symmetric family
This page was built for publication: Inadmissibility of the usual estimator for the location parameters of spherically symmetric distributions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2641017)
