Screening, Transforming, and Fitting Predictors for the Cumulative Logit Model - Bruce Lund, Magnify Analytics

The cumulative logit model is a logistic regression model where the target (or dependent variable) has 2 or more ordered levels. If there are only 2 levels, then the cumulative logit is the binary logistic model. Predictors for the cumulative logit model might be “NOD” (nominal, ordinal, discrete) where typically the number of levels is under 20. Alternatively, predictors might be “continuous” where the predictor is numeric and has many levels. This paper discusses methods that screen and transform both NOD and continuous predictors before the stage of model fitting. Once a collection of predictors has been screened and transformed, the paper discusses predictor variable selection methods for model fitting. One focus of this paper is determining when a predictor should be allowed to have unequal slopes. If unequal slopes are allowed, then the predictor has J-1 distinct slopes corresponding to the J values of the target variable. SAS® macros are presented which implement screening and transforming methods. Familiarity with PROC LOGISTIC is assumed.

About the Presenter

Bruce Lund is an independent consultant who specializes in predictive modeling. For the prior 15 years he was a consultant for OneMagnify of Detroit. Before OneMagnify, he was the customer database manager at Ford Motor Company and a mathematics professor at University of New Brunswick, Canada. He has a mathematics PhD from Stanford University. Bruce has presented at SAS Global Forum, SAS AnalyticsX, ASA CSP, and frequently at MWSUG and MSUG. He has used SAS for over 30 years.