OPTIMAL STEIN-RULE ESTIMATOR UNDER QUARTIC LOSS FUNCTION

Authors

  • M. ARASHI
  • Z. JAFARI

Keywords:

James-Stein estimator, Minimax estimator, Quartic loss function, Scale mixture of multivariate normal distributions

Abstract

In this approach, we construct the necessary and sufficient conditions in order a James-Stein
type estimator outperforms the minimax estimator of the mean. Particularly we consider
a class of scale mixture of multivariate normal distributions and derive the dominating
conditions under the quartic loss function. Some leading examples are also exhibited for
checking the efficiency of the proposed model and specifying theoretical computations.

Downloads

Issue

Vol. 46 No. 1 (2012)

Section

Articles