We establish a sufficient condition for the occurrence of nonresponse boundary solutions in a two-way square contingency table with nonignorable nonresponse. The condition depends only on the observed counts and thus it does not require computing the maximum likelihood estimates.

Integer-valued time series models make use of thinning operators for coherency in the nature of count data. However, the thinning operators make residuals unobservable and are the main difficulty in developing diagnostic tools for autocorrelated count data. In this regard, we introduce a new...

We introduce a novel method of sensitivity analysis to aid in assessing the missing-at-random assumption in a two-way contingency table with one supplemental margin. The procedure involves a set of response odds and nonresponse odds computed from the observed counts.