Linear Representation of M-Estimates of Multiple Regression Coefficients  

作　　者：陈希孺 

机构地区：[1]Graduate School,Academia Sinica,Beijing 100039,PRC

出　　处：《Science China Mathematics》1994年第2期162-177,共16页中国科学：数学（英文版）

基　　金：Project supported by the National Natural Science Foundation of China

摘　　要：This paper studies the strong and weak representation of M-estimates of multiple regression coefficients when the convexity condition is not assumed.The order of the remainder term,or its principal part,is accurate.Using the result,we obtain the convergence rate,the LIL and Berry-Esseen type bounds of the M-estimate.This paper studies the strong and weak representation of M-estimates of multiple regression coefficients when the convexity condition is not assumed.The order of the remainder term,or its principal part,is accurate.Using the result,we obtain the convergence rate,the LIL and Berry-Esseen type bounds of the M-estimate.

关 键 词：linear regression model Af-estimate CONSISTENCY law of ITERATED logarism Berry-Esseen BOUNDS 

分 类 号：O241[理学—计算数学]