Accession Number : AD0673237

Title :   INDEPENDENCE AND STATISTICAL REDUCTION FOR INVARIANT MODELS.

Descriptive Note : Technical rept.,

Corporate Author : WISCONSIN UNIV MADISON DEPT OF STATISTICS

Personal Author(s) : Bondar,James

Report Date : MAY 1968

Pagination or Media Count : 30

Abstract : A theorem is derived, proving the independence of minimal sufficient statistics and the 'natural' ancillary statistic for a class of invariant models. The theorem is applied to prove the equivalence of different methods of reducing statistical models, e.g. invariance reductions and sufficiency-followed-by-invariance reductions. (Author)

Descriptors :   (*STATISTICAL ANALYSIS, MATHEMATICAL MODELS), STATISTICAL DISTRIBUTIONS, GROUPS(MATHEMATICS), INVARIANCE, TRANSFORMATIONS(MATHEMATICS), MEASURE THEORY, TOPOLOGY, THEOREMS

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE