作者: S. Kullback , M. Kupperman , H. H. Ku
DOI: 10.1080/00401706.1962.10490041
关键词:
摘要: A number of useful tests for contingency tables and finite stationary Markov chains are presented in this paper based on the use notions information theory. consistent simple approach is used developing various test procedures results given form analysis-of-information tables. Beginning with hypotheses a one-way table, specified probabilities, independence, conditional homogeneity classifications, symmetry developed two, three, four, higher order classihcations. For chains, include matrix transition Markovity, several realizations chains. Worked examples throughout paper.