Minimax Shrinkage Estimators and Estimators Dominating the James-Stein Estimator under the Balanced Loss Function

Abdelkader Benkhaled, Abdenour Hamdaoui, Mekki Terbeche

2022

Full text Open access: Gold Metadata (CSL-JSON)

DOI 10.32523/2077-9879-2022-13-2-18-36

Data confidence 70 medium

How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This paper is dealing with the shrinkage estimators of a multivariate normal mean and their minimaxity properties under the balanced loss function. We present here two different classes of estimators: the first which generalizes the James-Stein estimator, and show that any estimator of this class dominates the maximum likelihood estimator (MLE), consequently it is minimax, and the second dominates the James-Stein estimator and we conclude that any estimator of this class is also minimax.