Optimal shrinkage estimator for high-dimensional mean vector

Taras Bodnar*, Ostap Okhrin, Nestor Parolya

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

14 Citations (Scopus)

Abstract

In this paper we derive the optimal linear shrinkage estimator for the high-dimensional mean vector using random matrix theory. The results are obtained under the assumption that both the dimension p and the sample size n tend to infinity in such a way that p∕n→c∈(0,∞). Under weak conditions imposed on the underlying data generating mechanism, we find the asymptotic equivalents to the optimal shrinkage intensities and estimate them consistently. The proposed nonparametric estimator for the high-dimensional mean vector has a simple structure and is proven to minimize asymptotically, with probability 1, the quadratic loss when c∈(0,1). When c∈(1,∞) we modify the estimator by using a feasible estimator for the precision covariance matrix. To this end, an exhaustive simulation study and an application to real data are provided where the proposed estimator is compared with known benchmarks from the literature. It turns out that the existing estimators of the mean vector, including the new proposal, converge to the sample mean vector when the true mean vector has an unbounded Euclidean norm.

Original languageEnglish
Pages (from-to)63-79
Number of pages17
JournalJournal of Multivariate Analysis
Volume170
DOIs
Publication statusPublished - 2019
Externally publishedYes

Keywords

  • Large-dimensional asymptotics
  • Mean vector estimation
  • Random matrix theory
  • Shrinkage estimator

Fingerprint

Dive into the research topics of 'Optimal shrinkage estimator for high-dimensional mean vector'. Together they form a unique fingerprint.

Cite this