摘要
在一般情形下,给出了在模M={Y,Xβ,σ2V}与删除第i个观测值后得到的模型Md={Yd,Xdβ,σ2Vd}下Xdβ的最佳线性无偏估计差的表达式.得到了二者相等的充要条件,给出了在模型Md下Xdβ的最小二乘估计是M下Xdβ的最佳线性无偏估计的充要条件,以及Md下σ2的最小范数二次无偏估计是M下σ2的最小范数二次无偏估计的充要条件.
A formula for the difference between the BLUEs of Xβ under M={Y,Xβ,σ 2V} and the BLUEs of X dβ under M d={Y d,X dβ,σ 2V d} is given.Further the necessary and sufficient conditions for the equality of above two estimators are established, and the necessary and sufficient conditions for the equality of the OLSE of X dβ under M d and the BLUEs of X dβ under M are established.It also give the necessary and sufficient condition for the NINQUE of σ 2 under M d to be the MINQUE of σ 2 under M.
出处
《东北师大学报(自然科学版)》
CAS
CSCD
1998年第3期36-40,共5页
Journal of Northeast Normal University(Natural Science Edition)
基金
国家自然科学基金
关键词
线性模型
广义逆
最小二乘估计
观测值
BLUE
linear model
generalied inverse
ordinary least squares estimator
best linear unbiased estimator
minimum norm quadratic unbiased estimator