LINEAR-MODELS WITHOUT MOMENTS
成果类型：
Article
署名作者：
JENSEN, DR
署名单位：
Virginia Polytechnic Institute & State University
刊物名称：
BIOMETRIKA
ISSN/ISSBN：
0006-3444
发表日期：
1979
页码：
611617
关键词：
摘要：
In models y = X.beta. + e having spherical errors without moments, the usual estimator .cxa..beta. is median-unbiased for .beta., most concentrated around .beta. among all median-unbiased linear estimates and consistent for .beta. in a sequence of identical but dependent experiments possibly having Cauchy errors. It is shown for all spherical error laws that the joint null distributions of variance ratios have exactly their normal theory forms, and for unimodal error laws that the power functions are monotone and the usual tests are unbiased.