Accession Number : AD0634774

Title :   GENERALIZED BAYES DECISION FUNCTIONS, ADMISSIBILITY AND THE EXPONENTIAL FAMILY.

Descriptive Note : Technical rept.,

Corporate Author : WISCONSIN UNIV MADISON DEPT OF STATISTICS

Personal Author(s) : Stone, M.

Report Date : MAY 1966

Pagination or Media Count : 12

Abstract : Three types of generalized Bayes decision functions NGBDF, EGBDF, CGBDF are defined and their relationships examined. A sufficient condition is given for the admissibility of a CGBDF. In the context of the exponential family and quadratic loss estimation, a sufficient condition that x/lambda +1)be a CGBDF for the mean is shown to be the same as Karlin's condition for the admissibility of x/(lambda +1). (Author)

Descriptors :   (*STATISTICAL FUNCTIONS, *DECISION THEORY), MEASURE THEORY

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE