Using Neural Networks for Credit Scoring

Abstract Neural networks offer an alternative to numerical scoring schemes for credit granting and extension decisions. A standard back‐ propagation neural network running on a DOS personal computer is used with 125 credit applicants whose loan outcomes are known. Applicant characteristics are described as input neurons receiving values representing the individuals' demographic and credit information. Three categories of payment history, delinquent, charged‐off, and paid‐off, are used as the networks output neurons to depict the loan outcomes. After training on part of the data, correct classifications were made on 76–80% of the holdout sample.