Acta mathematica scientia,Series A ›› 2013, Vol. 33 ›› Issue (4): 702-708.

• Articles • Previous Articles     Next Articles

A New Class of Estimators for Coefficients in Mixed Effect Linear Model

 XU Ying, HE Dao-Jiang*   

  1. School of Mathematics and Computer Science, Anhui Normal University, Anhui Wuhu 241000
  • Received:2012-04-19 Revised:2013-03-12 Online:2013-08-25 Published:2013-08-25
  • Contact: HE Dao-Jiang,djheahnu@163.com E-mail:djheahnu@163.com
  • Supported by:

    国家自然科学基金(11201005, 11271020)、安徽省高校自然科学研究基金重点项目(KJ2012A135)和安徽省高校优秀青年人才基金重点项目(2012SQRL028ZD)资助.

Abstract:

In the repeated-measures data model, in order to deal with the multicollinearity, a new class of estimator called s-K estimators for the parameters in mixed effect linear model are proposed. Under certain conditions, the new estimators are shown to be superior to the Stein estimators, Ridge estimators and least squared estimators, respectively.

Key words: Mixed effect, Linear model, s-K estimator, Stein estimator, Ridge estimator

CLC Number: 

  • 62F10
Trendmd