Power divergence statistics under quasi independence model for square contingency tables

Abstract

In incomplete contingency tables, some cells may contain structural zeros. The quasi-independence model, which is a generalization of the independence model, is most commonly model used to analyze incomplete contingency tables. Goodness of fit tests of the quasi-independence model are usually based on Pearson chi square test statistic and likelihood ratio test statistic. In power divergence statistics family, the selection of power divergence parameter is of interest in multivariate discrete data. In this study, a simulation study is conducted to evaluate the performance of the power divergence statistics under quasi independence model for particular power divergence parameters in terms of power values.