James-Stein type estimators of variances

Tiejun TONG, Homin Jang, Yuedong Wang*

*Corresponding author for this work

Research output: Contribution to journalJournal articlepeer-review

7 Citations (Scopus)

Abstract

In this paper we propose James-Stein type estimators for variances raised to a fixed power by shrinking individual variance estimators towards the arithmetic mean. We derive and estimate the optimal choices of shrinkage parameters under both the squared and the Stein loss functions. Asymptotic properties are investigated under two schemes when either the number of degrees of freedom of each individual estimate or the number of individuals approaches to infinity. Simulation studies indicate that the performance of various shrinkage estimators depends on the loss function, and the proposed estimator outperforms existing methods under the squared loss function.

Original languageEnglish
Pages (from-to)232-243
Number of pages12
JournalJournal of Multivariate Analysis
Volume107
DOIs
Publication statusPublished - May 2012

Scopus Subject Areas

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty

User-Defined Keywords

  • Inadmissibility
  • Shrinkage estimation
  • Shrinkage parameter
  • Squared loss function
  • Stein loss function
  • Variance estimation

Fingerprint

Dive into the research topics of 'James-Stein type estimators of variances'. Together they form a unique fingerprint.

Cite this