Consumer Credit-Risk Models Via Machine-Learning Algorithms
We apply machine-learning techniques to construct nonlinear nonparametric forecasting models of consumer credit risk. By combining customer transactions and credit bureau data from January 2005 to April 2009 for a sample of a major commercial bank’s customers, we are able to construct out-of-sample...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Elsevier B.V.
2011
|
| Online Access: | http://hdl.handle.net/1721.1/66301 https://orcid.org/0000-0003-4909-4565 https://orcid.org/0000-0003-2944-7773 |
| _version_ | 1826213615179399168 |
|---|---|
| author | Khandani, Amir Ehsan Kim, Adlar J. Lo, Andrew W. |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Khandani, Amir Ehsan Kim, Adlar J. Lo, Andrew W. |
| author_sort | Khandani, Amir Ehsan |
| collection | MIT |
| description | We apply machine-learning techniques to construct nonlinear nonparametric forecasting models of consumer credit risk. By combining customer transactions and credit bureau data from January 2005 to April 2009 for a sample of a major commercial bank’s customers, we are able to construct out-of-sample forecasts that significantly improve the classification rates of credit-card-holder delinquencies and defaults, with linear regression R2’s of forecasted/realized delinquencies of 85%. Using conservative assumptions for the costs and benefits of cutting credit lines based on machine-learning forecasts, we estimate the cost savings to range from 6% to 25% of total losses. Moreover, the time-series patterns of estimated delinquency rates from this model over the course of the recent financial crisis suggest that aggregated consumer credit-risk analytics may have important applications in forecasting systemic risk. |
| first_indexed | 2024-09-23T15:52:04Z |
| format | Article |
| id | mit-1721.1/66301 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T15:52:04Z |
| publishDate | 2011 |
| publisher | Elsevier B.V. |
| record_format | dspace |
| spelling | mit-1721.1/663012022-10-02T04:41:34Z Consumer Credit-Risk Models Via Machine-Learning Algorithms Khandani, Amir Ehsan Kim, Adlar J. Lo, Andrew W. Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Sloan School of Management Sloan School of Management. Laboratory for Financial Engineering Lo, Andrew W. Lo, Andrew W. Khandani, Amir Ehsan Kim, Adlar J. We apply machine-learning techniques to construct nonlinear nonparametric forecasting models of consumer credit risk. By combining customer transactions and credit bureau data from January 2005 to April 2009 for a sample of a major commercial bank’s customers, we are able to construct out-of-sample forecasts that significantly improve the classification rates of credit-card-holder delinquencies and defaults, with linear regression R2’s of forecasted/realized delinquencies of 85%. Using conservative assumptions for the costs and benefits of cutting credit lines based on machine-learning forecasts, we estimate the cost savings to range from 6% to 25% of total losses. Moreover, the time-series patterns of estimated delinquency rates from this model over the course of the recent financial crisis suggest that aggregated consumer credit-risk analytics may have important applications in forecasting systemic risk. Massachusetts Institute of Technology. Laboratory for Financial Engineering Massachusetts Institute of Technology. Center for Future Banking 2011-10-17T19:26:15Z 2011-10-17T19:26:15Z 2010-06 2010-05 Article http://purl.org/eprint/type/JournalArticle 0378-4266 http://hdl.handle.net/1721.1/66301 Khandani, Amir E., Adlar J. Kim, and Andrew W. Lo. “Consumer credit-risk models via machine-learning algorithms☆.” Journal of Banking & Finance 34 (2010): 2767-2787. https://orcid.org/0000-0003-4909-4565 https://orcid.org/0000-0003-2944-7773 en_US http://dx.doi.org/10.1016/j.jbankfin.2010.06.001 Journal of Banking and Finance Creative Commons Attribution-Noncommercial-Share Alike 3.0 http://creativecommons.org/licenses/by-nc-sa/3.0/ application/pdf Elsevier B.V. Lo |
| spellingShingle | Khandani, Amir Ehsan Kim, Adlar J. Lo, Andrew W. Consumer Credit-Risk Models Via Machine-Learning Algorithms |
| title | Consumer Credit-Risk Models Via Machine-Learning Algorithms |
| title_full | Consumer Credit-Risk Models Via Machine-Learning Algorithms |
| title_fullStr | Consumer Credit-Risk Models Via Machine-Learning Algorithms |
| title_full_unstemmed | Consumer Credit-Risk Models Via Machine-Learning Algorithms |
| title_short | Consumer Credit-Risk Models Via Machine-Learning Algorithms |
| title_sort | consumer credit risk models via machine learning algorithms |
| url | http://hdl.handle.net/1721.1/66301 https://orcid.org/0000-0003-4909-4565 https://orcid.org/0000-0003-2944-7773 |
| work_keys_str_mv | AT khandaniamirehsan consumercreditriskmodelsviamachinelearningalgorithms AT kimadlarj consumercreditriskmodelsviamachinelearningalgorithms AT loandreww consumercreditriskmodelsviamachinelearningalgorithms |