Credit Scorecards Research Articles

Not If Affordability Data Adds Value But How to Add Real Value by Leveraging Affordability Data: Enhancing Predictive Capability of Credit Scoring Using Affordability Data

The aim of this paper is to focus on the role of affordability of loans on credit risk at 3 different levels. The first being short term practical approach to optimally using affordability data to improve current credit scorecards by showing how Random Forests can be used to improve logistic regression. After this discussion we step back and look at how affordability fits into ensuring systemic risk and use the credit union industry as a base case model for using affordability measures to reduce systemic risk. Finally we turn to the question of whether credit risk is a source of sustainable competitive advantage and what measures would create greater welfare for the financial system as whole. It is the belief of the authors that proprietary black box credit systems and increasing complexity of unpredictable factors make it necessary to build an open credit systems viewpoint. Using this weltanschauung we conclude with a simple example of restructuring mortgage products using affordability and borrower well being as central to sustainable credit risk management. The article is comprised of survey and quantitative analysis showing how affordability has been used in the credit industry or not used. The article shows that credit decisions can be improved by leveraging affordability optimally and then also discusses policy implications of using affordability to ensure soundness and well being of consumers instead of simply determining whether the consumer will repay the loan successfully.

Credit scorecard based on logistic regression with random coefficients

Many credit scoring techniques have been used to build credit scorecards. Among them, logistic regression model is the most commonly used in the banking industry due to its desirable features (e.g., robustness and transparency). Although some new techniques (e.g., support vector machine) have been applied to credit scoring and shown superior prediction accuracy, they have problems with the results interpretability. Therefore, these advanced techniques have not been widely applied in practice. To improve the prediction accuracy of logistic regression, logistic regression with random coefficients is proposed. The proposed model can improve prediction accuracy of logistic regression without sacrificing desirable features. It is expected that the proposed credit scorecard building method can contribute to effective management of credit risk in practice.

Assessing naïve Bayes as a method for screening credit applicants

The naïve Bayes rule (NBR) is a popular and often highly effective technique for constructing classification rules. This study examines the effectiveness of NBR as a method for constructing classification rules (credit scorecards) in the context of screening credit applicants (credit scoring). For this purpose, the study uses two real-world credit scoring data sets to benchmark NBR against linear discriminant analysis, logistic regression analysis, k-nearest neighbours, classification trees and neural networks. Of the two aforementioned data sets, the first one is taken from a major Greek bank whereas the second one is the Australian Credit Approval data set taken from the UCI Machine Learning Repository (available at http://www.ics.uci.edu/~mlearn/MLRepository.html). The predictive ability of scorecards is measured by the total percentage of correctly classified cases, the Gini coefficient and the bad rate amongst accepts. In each of the data sets, NBR is found to have a lower predictive ability than some of the other five methods under all measures used. Reasons that may negatively affect the predictive ability of NBR relative to that of alternative methods in the context of credit scoring are examined.

The use of credit scoring in the mortgage industry

In times of increased focus on risk management, acquiring or growing comparatively low risk mortgage portfolios has become an attractive value proposition. Banks that pursue an aggressive growth strategy in this sector, do, however, require risk control mechanisms that enable them to make a clear judgment on how great a growth appetite they can afford to have in order to still grow profitably. Moreover, under Basel II, the proper quantification of mortgage portfolio risk tends to help the release of own capital, because the mortgage portfolio is one of those portfolios where the relative benefits of internal ratings-based approaches compared with the standardised approach are greatest. Credit scoring models in general, and credit scorecards in particular, are suitable methods for quantifying the risk of an individual mortgage applicant or mortgage customer. In addition to score card development, this paper reviews alternative scoring model types that could be used for mortgage scoring. It presents reasons why it is beneficial to build such models in-house, before focusing on the steps necessary for building a mortgage scorecard. Finally, it discusses the important topics of creating segments, deploying models and eventually monitoring models.