THE ESTIMATE OF THE RANK FOR REGRESSION COEFFICIENT MATRIX IN A MEDIAN REGRESSION MODEL

作     者：缪柏其 

作者机构：Department of Mathematics University of Science and Technology of China Hefei Anhui 230026 China 

出 版 物：《Chinese Annals of Mathematics,Series B》 (数学年刊（B辑英文版）)

年 卷 期：1992年第13卷第2期

页      面：205-214页

核心收录：

学科分类：07[理学] 0701[理学-数学] 070101[理学-基础数学] 

基　　金：Projects supported by the National Science Foundation of China 

主　　题：multivariate median satisfy criterion matrices infer hereafter Borel lemma underlying 

摘      要：It is discussed to infer the rank of regression coefficient matrix in a multivariate linear regression model. If the zero median vector is unique and the design matrices satisfy some weaker conditions, it is derived that the estimators of the rank of regression coefficient matrix under the minimum distance criterion by using model selection method is strongly consistent.