ANALYSIS OF CONTINGENCY-TABLES BY CORRESPONDENCE MODELS SUBJECT TO ORDER CONSTRAINTS

Inferential correspondence analysis, which has gained much attention i n recent years, is applied here to contingency tables with ordered cat egories. To reflect such order, the parameters of the underlying corre spondence models are constrained to follow the order induced by the ca tegories of the analyzed table. A reparameterization of the correspond ence model in terms of a latent variable model is presented. This allo ws a simple and straightforward use of the EM algorithm to obtain effi cient order-restricted estimates. A goodness-of-fit test is also discu ssed, and an example is analyzed. A small Monte Carlo example is prese nted.