Accession Number : AD0704797

Title :   A BAYES SOLUTION FOR THE PROBLEM OF RANKING POISSON PARAMETERS.

Descriptive Note : Technical rept.,

Corporate Author : SOUTHERN METHODIST UNIV DALLAS TEX DEPT OF STATISTICS

Personal Author(s) : Dixon,D. O. ; Bland,R. P.

Report Date : 13 MAR 1970

Pagination or Media Count : 12

Abstract : The Poisson distribution is used as the mathematical model for the random behavior of many natural phenomena. For example, it is often used in reliability studies for the number of failures in some fixed time interval. The Poisson distribution (or population) is a one-parameter family of distributions, the parameter being the mean value. This paper is concerned with the problem of ranking the parameters for n different Poisson populations based on fixed equal size samples from each population. A Bayes solution is derived for several types of loss functions and gamma priors, under the usual assumptions of symmetric and additive losses and symmetric priors. (Author)

Descriptors :   (*STATISTICAL DISTRIBUTIONS, DECISION THEORY), PROBABILITY DENSITY FUNCTIONS, MATHEMATICAL MODELS, RELIABILITY, SAMPLING, THEOREMS

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE