ABSTRACT
This chapter begins a discussion by defining the typical two-group (binary) logistic regression model. Logistic regression analysis is a recognized technique for classifying individuals into two groups. Perhaps less known but equally important, polychotomous logistic regression (PLR) analysis is another method for performing classification. The chapter presents PLR analysis as a multigroup classification technique. It illustrates the technique using a cellular phone market segmentation study to build a market segmentation classification model as part of a customer relationship management (CRM) strategy. The cellular carrier provides its billing record information for appending to the survey. The chapter also presents a brief review of the estimation and modeling processes used in PLR. The estimation method for PLR is maximum likelihood estimation, the same method for binary logistic regression (BLR) estimation. The theory of stepwise variable selection, model assessment, and validation has been worked out for PLR.
