Scoring by usage

This paper aims to discover whether the predictive accuracy of a new applic ant scoring model for a credit card can be improved by estimating separate scoring models for applicants who are predicted to have high or low usage o f the card. Two models are estimated. First we estimate a model to explain the desired usage of a card, and second we estimate separately two further scoring models, one for those applicants whose usage is predicted to be hig h, and one for those for whom it is predicted to be low. The desired usage model is a two-stage Heckman model to take into account the fact that the o bserved usage of accepted applicants is constrained by their credit limit. Thus a model of the determinants of the credit limit, and one of usage, are both estimated using Heckman's ML estimator. We find a large number of var iables to be correlated with desired usage. We also find that the two stage scoring methodology gives only very marginal improvements over a single st age scoring model, that we are able to predict a greater percentage of bad payers for low users than for high users and a greater percentage of good p ayers for high users than for low users.