Show all articles
How to Measure the Quality of Credit Scoring Models
JEL classification:
C10, C53, D81, G32
Keywords:
credit scoring, quality indices, lift, profit, normally distributed scores
Abstract
Credit scoring models are widely used to predict the probability of client default. To measure the quality of such scoring models it is possible to use quantitative indices such as the Gini index, Kolmogorov-Smirnov statistics (KS), Lift, the Mahalanobis distance, and information statistics. This paper reviews and illustrates the use of these indices in practice.