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

一般线性模型下方差最小范数二次无偏估计的稳健性
引用本文:李树有,张宝学.一般线性模型下方差最小范数二次无偏估计的稳健性[J].数学研究及应用,2004,24(2):280-284.
作者姓名:李树有  张宝学
作者单位:1. 辽宁工学院数理系,辽宁,锦州,121001
2. 北京理工大学数学系,北京,100081;东北师范大学统计系,吉林,长春,130024
基金项目:Supported by China Mathematics Tian Yuan Youth Foundation (10226024) and China Postdoctoral Science Foundation.
摘    要:本文给出了一般线性模型下,由最小二乘得到的方差估计与最小范数二次无偏估计相等的充分必要条件,并且当Gauss-Markov估计与最小二乘估计相等时,可以得到一个简单的等价条件。

关 键 词:一般线性模型    广义逆    正交投影    最小范数二次无偏估计
收稿时间:5/5/2001 12:00:00 AM

Robustness of Minimum Norm Quadratic Unbiased Estimator of Variance for the General Linear Model
LI Shu-you and ZHANG Bao-xue.Robustness of Minimum Norm Quadratic Unbiased Estimator of Variance for the General Linear Model[J].Journal of Mathematical Research with Applications,2004,24(2):280-284.
Authors:LI Shu-you and ZHANG Bao-xue
Institution:Dept. of Math. Phys.; Liaoning Institute of Technology; Jinzhou; China;Dept. of Math; Beijing Institute of Technology; Beijing; China; Dept. of Statistics; Northeast Normal University Jilin; China
Abstract:In this paper, neccssary and sufficient conditions for equalities between a2y′(I-Px)y and ( )2 under the gcneral linear model, where ( )2 y′T1-2+(I-PT1/2+x)T1-2+y = rankT-rankX and a2 is a known positive number, are derived. Furthermore, when the Gauss-Markov estimators and the ordinary lcast squares estimators are identical, we obtain a simple equivalent condition.
Keywords:general linear model  generalized inverse  orthogonal projector  minimum norm quadratic unbiased estimator
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《数学研究及应用》浏览原始摘要信息
点击此处可从《数学研究及应用》下载免费的PDF全文
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