NUMERICAL TABULATION OF THE DISTRIBUTION OF KOLMOGOROVS STATISTIC FOR FINITE SAMPLE SIZE
成果类型：
Article
署名作者：
BIRNBAUM, ZW
刊物名称：
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
ISSN/ISSBN：
0162-1459
DOI：
10.2307/2281313
发表日期：
1952
页码：
425-441
关键词：
摘要：
Tables for the Kolmogorov-Smirnov test for goodness of fit given previously by Massey, one of whose papers was reviewed in these abstracts (35632), have been recomputed and expanded. The tables may be used to compute confidence intervals for the true distribution and to estimate sample sizes required for confidence intervals of prespecified width. It is shown that for N > 80, the asymptotic approximation is good enough for practical purposes. The Kolmogorov-Smirnov method is compared with other distribution-free methods for testing goodness of fit.