首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Admissibility of linear estimators of regression coefficients under quadratic loss
Authors:Jinlong Zhan  Chen Jianbao
Institution:(1) Kunming Institute of Technology, 650093 Kunming, China;(2) Yunnan University, 650091 Kunming, China
Abstract:For the general fixed effects linear model:Y=Xtau+epsi, epsi simN(0,V),Vge0, we obtain the necessary and sufficient conditions forLY+a to be admissible for a linear estimable functionStau in the class of all estimators under the loss function (d -Stau)primeD(d -Stau), whereDge0 is known. For the general random effects linear model: 
$$Y = X\beta + \varepsilon ,(\begin{array}{*{20}c} \beta \\ \varepsilon \\ \end{array} ) \sim N((\begin{array}{*{20}c} {A\alpha } \\ 0 \\ \end{array} ),(\begin{array}{*{20}c} {V_{11} } & {V_{12} } \\ {V_{21} } & {V_{22} } \\ \end{array}$$
Lambda=XV 11 Xprime+XV 12+V 21 Xprime+V 22ge0, we also get the necessary and sufficient conditions forLY+a to be admissible for a linear estimable functionSagr+Qbeta in the class of all estimators under the loss function (d -Sagr -Qbeta)primeD(d -Sagr -Qbeta), whereDge0 is known.
Keywords:
本文献已被 SpringerLink 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号