Author(s)

N. Meeus, J. Huysmans, B. Baesens, J. Vanthienen & M. Vandebroek

Abstract

This paper discusses the use of knowledge discovery techniques for a recent development in the field of scoring: behavioural scoring. The goal of behavioural scoring is to develop a model that predicts the creditworthiness of existing customers on the basis of their behaviour in the past. This paper explains briefly the Knowledge Discovery in Data process and applies the technique of logistic regression to real life datasets of a Belgian financial institution. It describes the development of scoring models for a cheque account, a credit account and the customer level and compares the model results for different pre-processing values and selection methods by means of the ROC curve, p-values and misclassification rates. Keywords: behavioural scoring, ROC-analysis, credit scoring, KDD. 1 Introduction Credit companies have always aimed at predicting as accurate as possible the creditworthiness of their (future) customers. In the past this was based on the personal judgement of the lender. Nowadays however credit-granting decisions are based on statistical or operational research methods. The continuous search for better techniques led to the idea of applying Knowledge Discovery in Data (KDD) techniques for scoring. A financial institution has access to lots of customer data in which one can possibly find interesting patterns and correlations that can contribute to a better credit granting. To indicate the prediction of customer creditworthiness one often uses the term ‘scoring’. Through the help of a scoring technique the customer is given a

Keywords

behavioural scoring, ROC-analysis, credit scoring, KDD.