An Evaluation of Alternative Credit Scoring Models in the Egyptian Banking Sector

Credit scoring is regarded as a one the most important techniques in banks and has become a very critical
tool during recent decades. A number of credit scoring models have been developed to evaluate the credit
risk process of both new and existing loan clients. This research aims to investigate the ability of neural nets,
such as probabilistic neural nets and multi-layer feed-forward nets, as well as conventional techniques, such
as the weight of evidence measure, discriminant analysis, probit analysis, and logistic regression as credit
scoring statistical tools. Results so far have revealed that neural nets give better average correct
classification rates than the other traditional techniques. In this research, analyses of the credit scoring task
are performed on one bank’s personal loans’ data-set. In order to evaluate the misclassification costs, both
type I errors and type II errors have been calculated.