English

On admissible estimation of a mean vector when the scale is unknown

Statistics Theory 2021-02-25 v1 Statistics Theory

Abstract

We consider admissibility of generalized Bayes estimators of the mean of a multivariate normal distribution when the scale is unknown under quadratic loss. The priors considered put the improper invariant prior on the scale while the prior on the mean has a hierarchical normal structure conditional on the scale. This conditional hierarchical prior is essentially that of Maruyama and Strawderman (2021, Biometrika) (MS21) which is indexed by a hyperparameter aa. In that paper aa is chosen so this conditional prior is proper which corresponds to a>1a>-1. This paper extends MS21 by considering improper conditional priors with aa in the closed interval [2,1][-2, -1], and establishing admissibility for such aa. The authors, in Maruyama and Strawderman (2017, JMVA), have earlier shown that such conditional priors with a<2a < -2 lead to inadmissible estimators. This paper therefore completes the determination of admissibility/inadmissibility for this class of priors. It establishes the the boundary as a=2a = -2, with admissibility holding for a2a\geq -2 and inadmissibility for a<2a < -2. This boundary corresponds exactly to that in the known scale case for these conditional priors, and which follows from Brown (1971, AOMS). As a notable benefit of this enlargement of the class of admissible generalized Bayes estimators, we give admissible and minimax estimators in all dimensions greater than 22 as opposed to MS21 which required the dimension to be greater than 44. In one particularly interesting special case, we establish that the joint Stein prior for the unknown scale case leads to a minimax admissible estimator in all dimensions greater than 22.

Cite

@article{arxiv.2102.12079,
  title  = {On admissible estimation of a mean vector when the scale is unknown},
  author = {Yuzo Maruyama and William E. Strawderman},
  journal= {arXiv preprint arXiv:2102.12079},
  year   = {2021}
}

Comments

32 pages, 1figure

R2 v1 2026-06-23T23:27:41.479Z