EFFECT OF A MISSING OBSERVATION ON HOTELLING GENERALIZED T02

作     者：KSHIRSAGAR, AM 

作者机构：Biostatistics Department University of Michigan Ann Arbor MI 48109 USA 

出 版 物：《STATISTICS & PROBABILITY LETTERS》 (统计学与概率论通讯)

年 卷 期：1989年第8卷第2期

页      面：163-166页

核心收录：

学科分类：07[理学] 0714[理学-统计学（可授理学、经济学学位）] 0701[理学-数学] 070101[理学-基础数学] 

基　　金：University of Manitoba Natural Sciences and Engineering Research Council of Canada 

主　　题：missing observation Hotelling's generalized T 0 2 linear combination of Wishart matrices bias in hypothesis matrix multivariate linear model 

摘      要：A generalization of the missing plot technique in univariate linear models to the multivariate case is given in this paper. The estimate of the missing observation vector produces the correct error matrix and the correct parameter estimates but the hypothesis matrix is biased. The adjustment required to the test, for achieving the desired level of significance, when the biased hypothesis matrix is used, is derived.