作者: William G. Cochran
关键词: Minimum distance estimation 、 Expected value 、 Subdivision 、 Chi-square test 、 Likelihood-ratio test 、 Statistics 、 Statistical theory 、 Goodness of fit 、 Mathematics 、 Test (assessment)
摘要: This paper contains an expository discussion of the chi square test goodness fit, intended for student and user statistical theory rather than expert. Part I describes historical development distribution on which rests. Research bearing practical application test--in particular minimum expected number per class construction classes--is discussed in II. Some varied opinions about extent to actually is useful scientist are presented III. IV outlines a tests that have been proposed as substitutes (the $\omega^2$ test, smooth likelihood ratio test) V supplementary run based low moments, subdivision into components).