摘要
本文证明了线性回归中协方差阵扰动模型岭估计的某种极限是数据删除模型的岭估计，建立了协方差阵扰动模型岭估计与Ｇ－Ｍ模型岭估计之间的关系，讨论了协方差阵扰动和数据删除对岭估计的混合影响，求得了岭估计的基于有偏估计的Ｃｏｏｋ距离。
Abstract In this paper,we prove that some limit of Ridge Estimate β(k)G.in the regression model with covariance matrix disturbance is Ridge Estimate β(k) in the G-M model. Next,we establish the relationships between β(k)G and β(k) and study the influence analysis on Ridge Estimate in the mixed regression model with variance disturbance and date-delection. And then, we give Cook distance (based on biased estimate) of Redge Estimate.
出处
《工程数学学报》
CSCD
1995年第3期83-88,共6页
Chinese Journal of Engineering Mathematics
