PDF Ebook Validation of Consumer Credit Risk Models
On November 19, 2004, the Payment Cards Center of the Federal Reserve Bank of Philadelphia, in conjunction with the Wharton School’s Financial Institutions Center, hosted a one-day event entitled “Forum on Validation of Consumer Credit Risk Models.” This forum brought together experts from industry, academia, and the policy community to discuss challenges surrounding model validation strategies and techniques. This paper provides highlights from the forum and ensuing discussion.
On November 19, 2004, the Payment Cards Center of the Federal Reserve Bank of Philadelphia and the Wharton School’s Financial Institutions Center hosted a “Forum on Validation of Consumer Credit Risk Models.”1 This one-day event brought together experts from industry, academia, and the policy community to discuss challenges surrounding model validation strategies and techniques. The discussions greatly benefited from the diverse perspectives of conference participants and the leadership provided by moderators and program speakers.
Retail lenders, and particularly credit card lenders, use statistical models extensively to guide a wide range of decision processes associated with loan origination, account management, and portfolio performance analysis. The increased sophistication of modeling techniques and the broader application of models have undoubtedly played key roles in the rapid growth of the credit card industry and consumer lending in general.3 At the same time, the widespread adoption of statistical modeling in these business processes has introduced new risk management challenges. Very simply, how do we know that our credit risk models are working as intended?
The conference discussions focused on two critical types of risk models: credit scoring models commonly used in credit underwriting and loss forecasting models used to predict losses over time at the portfolio level. These two model types differ in a number of ways, but the two modeling processes have strong theoretical links (although they are not often linked in practice).
Credit scoring models used for acquiring accounts are typically built on a static sample of accounts for which credit bureau — and often other applicant or demographic — information is available at the time of application. These data must then be combined with information about how these accounts ultimately performed in their first one to two years after acquisition. Credit scoring models are designed to predict the probability that an individual account will default or, more generally, develop a delinquency status bad enough that the bank would not have booked the account initially had it known this would happen. A number of credit scoring models only use credit bureau data to predict this probability, while others use application or demographic data in addition to credit bureau data.
Download
PDF Ebook Validation of Consumer Credit Risk Models
Posted in :
