
Accession Number : AD0462032
Title : OPTIMAL ROBUSTNESS FOR ESTIMATORS AND TESTS.
Descriptive Note : Technical rept.,
Corporate Author : NEW YORK UNIV NY COURANT INST OF MATHEMATICAL SCIENCES
Personal Author(s) : Birnbaum, Allan ; Laska, Eugene
Report Date : FEB 1965
Pagination or Media Count : 55
Abstract : The present paper is intended to complement research in efficiencyrobustness of estimators, by supplying formulations of concepts, techniques, and initial results for optimally efficiencyrobust estimators and tests in several types of problems. The present approach may be described as a formal indexing of alternative specifications (e.g. shapes of errordistributions) by a nuisance parameter, and adaptation of admissibility and related concepts and Bayes techniques of the NeymanPearson and Wald theories to the estimation and testing problems thus formulated. Specific problemsfor which new optimal efficiencyrobust estimators are given are: linear estimation of location parameters; rank tests and related estimators for twosample problems; and unbiased estimation. A byproduct is a generalization of Stein's characterization of locallybest unbiased estimators to the class of admissible unbiased estimators together with the corresponding complete class theorem. (Author)
Descriptors : (*STATISTICAL TESTS, RELIABILITY), PROBABILITY, MATHEMATICAL PREDICTION, THEOREMS, LINEAR SYSTEMS, STATISTICAL FUNCTIONS, CONVEX SETS, FUNCTIONAL ANALYSIS, SEQUENCES(MATHEMATICS), INTEGRALS, BIBLIOGRAPHIES.
Distribution Statement : APPROVED FOR PUBLIC RELEASE