Measures of Fit for Multinomial Discrete Models

In a discrete model, the predicted probabilities of a particular event can be matched to the observed (0, I) outcomes and this will give rise to a measure of fit for that event. Previous results for the binomial model are applied to multinomial models. In these models the measure of fit will vary between the various events, indicating that the model performs better for some than for others. In addition to these differential measures of fit for each separate event a single overall measure is constructed for the model as a whole. <BR><BR>