ABSTRACT
This chapter aims to develop tools for converting the types of predictors into a form that a model can better utilize. It provides approaches for and illustrates how to handle continuous predictors with commonly occurring issues. The predictors may: be on vastly different scales; contain a small number of extreme values; and be censored on the low and/or high end of the range. The chapter also illustrates methods for expanding individual predictors into many predictors in order to better represent more complex predictor-response relationships and to enable the extraction of predictive information. There are a variety of modifications that can be made to an individual predictor that might improve its utility in a model. Transformations can be made on a single numeric predictor to expand it to many predictors. The predictors are usually any measurements that may have a remote chance of potentially being related the response.
