Improved minimax estimation of the bivariate normal precision matrix under the squared loss

Xiaoqian Sun*, Xian Zhou

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

Suppose that n independent observations are drawn from a multivariate normal distribution Np (μ, Σ) with both mean vector μ and covariance matrix Σ unknown. We consider the problem of estimating the precision matrix Σ- 1 under the squared loss L (over(Σ, ^)- 1, Σ- 1) = tr (over(Σ, ^)- 1 Σ - Ip)2. It is well known that the best lower triangular equivariant estimator of Σ- 1 is minimax. In this paper, by using the information in the sample mean on Σ- 1, we construct a new class of improved estimators over the best lower triangular equivariant minimax estimator of Σ- 1 for p = 2. Our improved estimators are in the class of lower-triangular scale equivariant estimators and the method used is similar to that of Stein [1964. Inadmissibility of the usual estimator for the variance of a normal distribution with unknown mean. Ann. Inst. Statist. Math. 16, 155-160.].

Original languageEnglish
Pages (from-to)127-134
Number of pages8
JournalStatistics and Probability Letters
Volume78
Issue number2
DOIs
Publication statusPublished - 1 Feb 2008
Externally publishedYes

Keywords

  • Best lower triangular equivariant minimax estimator
  • Bivariate normal distribution
  • Inadmissibility
  • Precision matrix
  • The squared loss

Fingerprint

Dive into the research topics of 'Improved minimax estimation of the bivariate normal precision matrix under the squared loss'. Together they form a unique fingerprint.

Cite this