The performance of the adaptive optimal estimator under the extended balanced loss function

ÖZBAY N., KAÇIRANLAR S.

COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, vol.46, no.22, pp.11315-11326, 2017 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 46 Issue: 22
  • Publication Date: 2017
  • Doi Number: 10.1080/03610926.2016.1267760
  • Journal Name: COMMUNICATIONS IN STATISTICS-THEORY AND METHODS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.11315-11326
  • Keywords: Adaptive optimal estimator, Extended balanced loss function, Large sample properties, Linear regression model, Ordinary least squares estimator, SQUARED ERROR ESTIMATOR, STEIN-RULE ESTIMATOR, REGRESSION-COEFFICIENTS, OPERATIONAL VARIANTS, LINEAR-REGRESSION
  • Çukurova University Affiliated: Yes

Abstract

The adaptive optimal estimator of Farebrother (1975) is discussed by many authors, but the goodness of fitted model criterion that is used to investigate the performance of estimators is quite often ignored. Shalabh, Toutenburg, and Heumann (2009) proposed the extended balanced loss function in which the mean squared error and the Zellner's balanced loss function are just special cases of it. In this paper, we discuss the performance of the adaptive optimal estimator of Fare-brother (1975) under the extended balanced loss function. Moreover, a Monte Carlo simulation experiment is conducted to examine the performance of the estimator in finite samples.