OBTAINING THE MAXIMUM-LIKELIHOOD-ESTIMATES IN INCOMPLETE R X C CONTINGENCY-TABLES USING A POISSON GENERALIZED LINEAR-MODEL

This article describes estimation of the cell probabilities in an R x C contingency table with ignorable missing data. Popular methods for m aximizing the incomplete data likelihood are the EM-algorithm and the Newton-Raphson algorithm. Both of these methods require some modificat ion of existing statistical software to get the MLEs of the cell proba bilities as well as the variance estimates. We make the connection bet ween the multinomial and Poisson likelihoods to show that the MLEs can be obtained in any generalized linear models program without addition al programming or iteration loops.