International Biometric Society A mixture of experts model for rank data

A voting bloc is defined to be a group of voters who have similar voting preferences. The cleavage of the Irish electorate into voting blocs is of interest. Irish elections employ a ‘single transferable vote ’ electoral system; under this system voters rank some or all of the electoral candidates in order of preference. These rank votes provide a rich source of preference infor-mation from which inferences about the composition of the electorate may be drawn. Addition-ally, the influence of social factors or covariates on the electorate composition is of interest. A mixture of experts model is a mixture model in which the model parameters are functions of covariates. A mixture of experts model for rank data is developed to extend the current methodology to deal with the case of rank response data. The Benter model for rank data [1] is employed as the family of component densities within the mixture of experts model. Gen-eralized linear model theory is employed to model the influence of covariates on the mixing proportions within the mixture model. Model fitting is achieved via a hybrid EM/MM algorithm. The Irish presidential electorate is examined to provide an illustration of the mixture of experts model for rank data. The method provides a model-based clustering of the voters into voting blocs, examines the influence of social factors on this clustering and highlights the characteristi