Estimation in a linear regression model with stochastic linear restrictions: a new two-parameter-weighted mixed estimator


ÖZBAY N., KAÇIRANLAR S.

JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, vol.88, no.9, pp.1669-1683, 2018 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 88 Issue: 9
  • Publication Date: 2018
  • Doi Number: 10.1080/00949655.2018.1442836
  • Journal Name: JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.1669-1683
  • Keywords: Stochastic linear restrictions, two-parameter estimator, two-parameter-weighted mixed estimator, weighted mixed estimator, RIDGE-REGRESSION, BIASED-ESTIMATION
  • Çukurova University Affiliated: Yes

Abstract

The present paper considers the weighted mixed regression estimation of the coefficient vector in a linear regression model with stochastic linear restrictions binding the regression coefficients. We introduce a new two-parameter-weighted mixed estimator (TPWME) by unifying the weighted mixed estimator of Schaffrin and Toutenburg [1] and the two-parameter estimator (TPE) of ozkale and Kacranlar [2]. This new estimator is a general estimator which includes the weighted mixed estimator, the TPE and the restricted two-parameter estimator (RTPE) proposed by ozkale and Kacranlar [2] as special cases. Furthermore, we compare the TPWME with the weighted mixed estimator and the TPE with respect to the matrix mean square error criterion. A numerical example and a Monte Carlo simulation experiment are presented by using different estimators of the biasing parameters to illustrate some of the theoretical results.