POLYCHOTOMOUS REGRESSION

An automatic procedure that uses linear splines and their tensor produ cts is proposed for fitting a regression model to data involving a pol ychotomous response variable and one or more predictors. The fitted mo del can be used for multiple classification. The automatic fitting pro cedure involves maximum likelihood estimation, stepwise addition, step wise deletion, and model selection by the Akaike information criterion , cross-validation, or an independent test set. A modified version of the algorithm has been constructed that is applicable to large dataset s, and it is illustrated using a phoneme recognition dataset with 250, 000 cases, 45 classes, and 63 predictors.