The Method of Construction Scoring Cards Using SAS Platform
DOI:
https://doi.org/10.20535/1810-0546.2016.2.67487Keywords:
Risk management, Data mining, Credit scoring, Scoring card, Logistic regression, Classification qualityAbstract
Background. Development of effective methods for evaluating solvency of individuals and risk of banks in providing consumer loans.
Objective. Determining of the mechanisms for implementation of scoring models in the form of scoring cards. Analysis of the possibility of using scoring cards as a tool for credit risk management.
Methods. Construction of scoring cards and preliminary analysis of input data using specialized component of the SAS Enterprise Miner.
Results. The main stages of scoring cards development were considered. The scoring card was constructed that is based on actual statistical data on granting of the consumer loans. The research also presents comparative analysis of the scoring cards with other statistical methods of subjects classification.
Conclusions. It was established in this study that the scoring cards have better forecasting ability than other statistical methods such as decision trees, neural networks and logistic regression. The format of development the forecasting models in the form of scoring cards is the easiest for interpreting. However, application of this method requires considerable investments as well as continuous updating and renewal of credit histories for borrowers.References
T.R. Bielecki and M. Rutkowsky, Credit Risk: Modeling, Valuation, Hedging.Berlin,Germany: Springer, 2002.
H. van Gruening and S.B. Bratanovic, Analyzing and Managing Banking Risks.Washington: The World Bank, 2003.
T. Aven, Foundations of Risk Analysis: A Knowledge and Decision-Oriented Perspective.New York: John Wiley & Sons, Ltd., 2003.
G. Xuesong et al., “Corporate credit rating model using support vector domain combined with fuzzy clustering algorithm”, Math. Problems Eng., vol. 1, 2012, pp. 1–20.
T. Lunkina. (2015). Using Scoring Models in Consumer Lending Risk Management [Online]. Available: http://www.economy. nayka.com.ua/?op=1&z=3792 (in Ukrainian).
A. Sorokin. (2014). Building a Scorecard Using a Logistic Regression Model. [Online]. Available: http://naukovedenie.ru/PDF/ 180EVN214.pdf (in Russian).
N. Siddiki, Scorecards for Credit Risk Assessment. Development and Implementation of Intelligent Methods of Credit Scoring.Moscow,Russia, 2014 (in Russian).
A. Sorokin. (2014). On the Question of Validation of a Logistic Regression Model in Credit Scoring. [Online]. Available: http://naukovedenie.ru/PDF/ 173EVN214.pdf (in Russian).
A. Terentyev et al., SAS BASE: Programming Basics.Kyiv,Ukraine: Edelveis Publishers, 2015 (in Russian).
B.S. Anderson and R.W. Thompson, Developing Credit Scorecards Using SAS Credit Scoring for Enterprise Miner 5.3. Cary: SAS Institute Inc, 2009.
Downloads
Published
Issue
Section
License
Copyright (c) 2017 NTUU KPI Authors who publish with this journal agree to the following terms:- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under CC BY 4.0 that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work