Stein岭型主成分估计下多个数据删除模型的强影响分析  

Strong Impact Analysis of Multiple Data Delete Model Based on Stein Ridge and Principal Components Estimator

作　　者：朱宁[1] 严冠东 刘庆华[1] 

机构地区：[1]桂林电子科技大学数学与计算科学学院,广西桂林541004

出　　处：《汕头大学学报（自然科学版）》2015年第2期20-27,共8页Journal of Shantou University：Natural Science Edition

基　　金：桂林电子科技大学研究生创新项目(GDYCSZ201471)

摘　　要：基于Stein岭型主成分估计下研究多个数据删除模型,探讨数据删除模型估计量的有关性质,并给出了多个数据删除模型的CRi统计量、APi统计量、Di统计量的新表达式.Strong impact analysis of multiple data delete model based on stein ridge and principal components estimator is studied. A strong impact on the analysis model under biased estimator is proposed. The property of estimators of the data deletion model is also discussed. Besides, the new expressions of CRi, A Pi and Di based on multiple data delete model, are given.

关 键 词：Stein岭型主成分估计 多个数据删除模型 多个强影响点 诊断统计量 

分 类 号：O212.1[理学—概率论与数理统计]