DOI: 10.1080/01621459.1970.10481191
关键词:
摘要: Abstract Many authors have been concerned with contingency tables containing cells which are missing, a priori zero or otherwise specified. This article examines the problem of maximum likelihood estimation for such under “quasi-in-dependence” model. In particular, conditions provided to ensure existence unique nonzero estimates incomplete obtained by deleting missing cells, even when other contain counts due sampling variation.