分块线性模型中Bayes线性无偏估计的可加性

广东工业大学学报 ›› 2012, Vol. 29 ›› Issue (1): 64-66.

分块线性模型中Bayes线性无偏估计的可加性

1．广东工业大学应用数学学院, 广东广州 510520；2．浙江财经学院数学与统计学院，浙江杭州，310018

出版日期:2012-03-25 发布日期:2012-03-25
作者简介:邱红兵（1974-），男，副教授，主要研究方向为线性模型、矩阵分析.
基金资助:
国家自然科学基金资助项目(11171058)；浙江省自然科学基金资助项目(Y6110615)；国家社会科学基金资助项目(11CTJ008)

Additivity of the Bayes Linear Unbiased Estimator in the Partitioned Linear Model

1.Faculty of Applied Mathematics, Guangdong University of Technology, Guangzhou 510520，China;
2.Faculty of Mathematics and Statistics, Zhejiang University of Finance and Economics, Hangzhou 310018, China

Online:2012-03-25 Published:2012-03-25

摘要/Abstract

摘要： 得到了βi 在线性模型M=(y,X1β1+X2β2,T)下的Bayes线性无偏估计与其在小模型Mi=(y,Xiβi,T)下的Bayes线性无偏估计相等的充分必要条件,i=1,2；并考虑了Xiβi在小模型Mi下的Bayes线性无偏估计之和与Xβ在全模型M下的Bayes线性无偏估计之间的相等关系.


关键词: 分块线性模型; Bayes线性无偏估计;最佳线性无偏估计;可加性

Abstract: 

Equality of Bayes linear unbiased estimator for βi in linear model M and M1is discussed, and the necessary and sufficient conditions are derived for the Bayes linear unbiased estimator in the partitioned linear model 〖WTHX〗M=(y,X1β1+X2β2,T) to be the sum of the Bayes linear unbiased estimators under the two small models M1=(y,X1β1,T)〖WTBZ〗 and 〖WTHX〗M2=(y,X2β2,T)

Key words: partitioned linear model; Bayes linear unbiased estimator; best linear unbiased estimator; additivity

邱红兵1，罗季2. 分块线性模型中Bayes线性无偏估计的可加性[J]. 广东工业大学学报, 2012, 29(1): 64-66.

Qiu Hongbing , Luo Ji . Additivity of the Bayes Linear Unbiased Estimator in the Partitioned Linear Model[J]. Journal of Guangdong University of Technology, 2012, 29(1): 64-66.

参考文献

［1］王松桂, 史建红, 尹素菊, 等. 线性模型引论［M］. 北京: 科学出版社, 2004. 

［2］邱红兵, 罗季. 线性模型中F模型的稳健性［J］. 数学学报中文版. 2010, 53(2): 385-392.

［3］ Trenkler G，Wei L S. The Bayes estimator in a misspecified regression model ［J］. Test. 1996（5）: 113-123. 

［4］张伟平. 贝叶斯线性无偏估计［D］.安徽：中国科学技术大学统计与金融系，2005.

［5］ Wei L S, Zhang W P. The superiorities of Bayes linear minimum risk estimation in linear model ［J］. Communications in StatisticsTheory and Methods. 2007, 36: 917-926. 

［6］韦来生. 错误先验假定下回归系数Bayes估计的小样本性质［J］．应用概率统计. 2000, 6(1): 71-80．

［7］张伟平, 韦来生. 错误先验假设下Bayes线性无偏估计的稳健性［J］.应用概率统计,2007,23(1): 59-67.

［8］霍涉云, 张伟平, 韦来生. 一类线性模型参数的Bayes估计及其优良性［J］.中国科学技术大学学报.2007, 37(7):773-776.

［9］童楠, 韦来生. 单向分类方差分析模型中参数的Bayes估计及其优良性［J］. 中国科学技术大学学报, 2008,38(9): 10841088.

［10］ Puntanen S. Some further results related to reduced singular linear model ［J］. Communications in StatisticsTheory and Methods,1997,26:375-385.

［11］ Zhang Baoxue. A study of the equivalence of the BLUEs between a partitioned singular linear model and its reduced singular linear models ［J］. Acta Mathematica Sinica, English Series,2004,20(3):557-568.

［12］ Yongge Tian. On an additive decomposition of the BLUE in a multiplePartitioned linear model ［J］. J Multivariate Anal，2009,100:767-776.

［13］ Yongge Tian, Yoshio Takane. On sum decompositions of weighted leastsquares estimators for the partitioned linear model［J］. Communications in Statistics Theory and Methods,2008,37:55-69.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed