A new Liu-type estimator

KURNAZ F. S., Akay K. U.

STATISTICAL PAPERS, vol.56, no.2, pp.495-517, 2015 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 56 Issue: 2
  • Publication Date: 2015
  • Doi Number: 10.1007/s00362-014-0594-6
  • Journal Name: STATISTICAL PAPERS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.495-517
  • Istanbul University Affiliated: Yes

Abstract

The purpose of this paper is to introduce a new general Liu-type estimator which includes the ordinary least squares (OLS), ordinary ridge regression (ORR), Liu estimators and some estimators with two biasing parameters as special cases. Also, we investigate the superiority of the new Liu-type estimator to the OLS, ORR, Liu estimators and the estimators with two biasing parameters under the matrix mean squared error (MMSE) criterion. Furthermore, the results are illustrated both theoretically and graphically on the Portland cement dataset which is widely used in literature.