A Comparison of Data Mining Techniques for Credit Scoring in Banking: A Managerial Perspective

Abstract

Abstract. Credit scoring is a very important task for lenders to evaluate the loan applications they receive from consumers as well as for insurance companies, which use scoring systems today to evaluate new policyholders and the risks these prospective customers might present to the insurer. Credit scoring systems are used to model the potential risk of loan applications, which have the advantage of being able to handle a large volume of credit applications quickly with minimal labour, thus reducing operating costs, and they may be an effective substitute for the use of judgment among inexperienced loan offi cers, thus helping to control bad debt losses. This study explores the performance of credit scoring models using traditional and artifi cial intelligence approaches: discriminant analysis, logistic regression, neural networks and classifi cation and regression trees. Experimental studies using real world data sets have demonstrated that the classifi cation and regression trees and neural networks outperform the traditional credit scoring models in terms of predictive accuracy and type II errors.

The Impact of the Implementation of Business Intelligent to Support Decision-Making in the American Banking Industry: A Literature Review

“…The role of business intelligence in solving business problems and decision-making at the other areas is the result of the implementation of customer credit scoring model (Ince & Aktan, 2009). Credit score of customers is the most important activity to evaluate loan applications submitted by the customer.…”

Readjusting Trade-Offs Among Criteria in Internal Ratings of Credit-Scoring: An Empirical Essay of Risk Analysis in Mortgage Loans

“…However, because of the effects of the current economic conjuncture, regulators have been encouraging banks to develop more sophisticated risk models with the purpose of "better quantifying the financial risks they face and assigning the necessary economic capital" (Lopez, Saidenberg 2000: 152). From this standpoint, remarkable progress has occurred recently in terms of credit risk modeling (for a categorized literature review, see Altman, Saunders 1998;Crook et al 2007;Suhobokov 2007;Ince, Aktan 2009;Yu et al 2009;Wang et al 2011). Still, as recognized by many (e.g.…”

DRIP – Data Rich, Information Poor: A Concise Synopsis of Data Mining

“…The researchers applied each of four different types of credit card scoring, discriminant analysis, logistic regression, decision trees (C5; CART) and neural networks to evaluate banking customers for credit. The researchers found that CART had above average results in classifying, but neural networks provided the better overall results for credit scoring (Ince & Aktan, 2009). …”