Premium
Inference for the maximum cell probability under multinomial sampling
Author(s) -
Gelfand A. E.,
Glaz J.,
Kuo L.,
Lee T.M.
Publication year - 1992
Publication title -
naval research logistics (nrl)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.665
H-Index - 68
eISSN - 1520-6750
pISSN - 0894-069X
DOI - 10.1002/1520-6750(199202)39:1<97::aid-nav3220390108>3.0.co;2-o
Subject(s) - frequentist inference , multinomial distribution , inference , bayesian probability , statistics , point estimation , econometrics , computer science , statistical inference , bayesian inference , sampling (signal processing) , mathematics , artificial intelligence , filter (signal processing) , computer vision
This article investigates inference for p max , the largest cell probability in multinomial trials for the case of a small to moderate number of trials. Emphasis focuses on point and interval estimation. Both frequentist and Bayesian approaches are developed. The results of extensive simulation investigation are included as well as the analysis of a set of crime data for the city of New Orleans taken from the National Crime Survey.