ANALYZING BIVARIATE ORDINAL DATA USING A GLOBAL ODDS RATIO

A moment method is proposed for regression analysis of bivariate order ed categorical data using the global odds ratio as the measure of asso ciation. For modeling the margins, this method utilizes the stochastic ordering implicit in the data. This method allows for covariate effec ts in modeling the association between responses. An application of th e proposed model is illustrated using the ophthalmological data from t he Wisconsin Epidemiologic Study of Diabetic Retinopathy for identifyi ng risk factors among younger onset diabetics. This model is also exte nded to a longitudinal data setting with more than two repeated measur es.