摘要
对线性模型:Y=Xβ+ρ,E(ρ)=0,cov(ρ)=σ2G,G=diag(g1,…gn),探讨了单个数据点对σ2的估计σ2,β的估计β的广义方差|covβ|及Fisher信息阵:σ-2X′G-1X的影响,给出了度量准则和它的化简式及统计解释.
In this paper we consider the linear model:and research the influence of one data case on estimation of σ2,generalized variance,|covβ|and Fisher information matrix.We also obtain some measuring criterions,calculating formulas and statistical explanations
出处
《数学研究》
CSCD
1996年第3期98-102,共5页
Journal of Mathematical Study