Symmetry and unimodality in linear inference

[摘要] Distribution-free results beyond Gauss-Markov theory-re found under weak assumptions regarding the errors. Symmetry, unimodality, and location-scale families are studied in estimation; nonstandard versions of Gauss-Markov results are given; and distribution-free confidence sets are tightened under symmetry and unimodality of errors. Normal-theory approximate tests are seen to exhibit monotone power in certain classes of symmetric unimodal errors. (C) 1997 Academic Press.

[发布日期] 1997-02-01

[关键词] symmetry under reflections;unimodal mixtures;location, median, and modal-unbiasedness;Gauss-Markov theory;tightened confidence bounds