On sum decompositions of weighted least-squares estimators for the partitioned linear model

被引：19
|
作者：
Tian, Yongge [1 ]
Takane, Yoshio [2 ]
机构：
[1] Shanghai Univ Finance & Econ, Sch Econ, Shanghai 200433, Peoples R China
[2] McGill Univ, Dept Psychol, Montreal, PQ, Canada
来源：
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS | 2008年 / 37卷 / 01期
基金：
加拿大自然科学与工程研究理事会;
关键词：
matrix rank method; Moore-Penrose inverse; partitioned linear model; projector; small models; sum decompositions; WLSE;
D O I：
10.1080/03610920701648862
中图分类号：
O21 [概率论与数理统计]; C8 [统计学];
学科分类号：
020208 ; 070103 ; 0714 ;
摘要：
For the partitioned linear model M={y, X-1 beta(1)+X-2 beta(2),sigma(2)Sigma}, this article investigates decompositions of weighted least-squares estimator (WLSE) of X-1 beta(1)+X-2 beta(2) under M as sums of WLSEs under the two small models {y, X-1 beta(1),sigma(2)Sigma} and {y, X-2 beta(2),sigma(2)Sigma}. Some consequences on the sum decomposition of the unique best unbiased linear estimator (BLUE) of X-1 beta(1)+X-2 beta(2) under M are also given.
引用
收藏
页码：55 / 69
页数：15
相关论文
共 50 条
12345