摘要
本文在一般的Gauss——Markov模型之下讨论了可估参数函数的线性充分统计量、最小线性充分统计量的稳健性以及线性充分统计量、最小线性充分统计量的稳健性,给出了相应的充要条件。
Necessary and Sufficient Conditions are derived for the class μ_0 of linearly sufficient statistics,the cless μ_0~* of linaerly minimal Sufficent statistics with respect to a given vector of parametric functions,the class of linearrly sufficient statistics,the class ~* of linearly minimal sufficient statistics,corresponding to a General Gauss-Markov Model M_0={Y,X_0β_0,V_0}to remain valid under an alternatve Model M={Y,Xβ,V}in the sense that μ_0μ,μ_0~*μ~*,,~*~* wher μ,μ~*,,~* are the counterparts of μ_0,μ_0~*,_0,_0~* corresponding to the Model M.
出处
《东北师大学报（自然科学版）》
CAS
CSCD
1992年第1期13-18,共6页
Journal of Northeast Normal University(Natural Science Edition)
关键词
G-M模型
充分统计量
鲁棒性
General Gauss-Markov Model
Linearly sufficient statistics
linearly minimal sufficient statistics
