Track 1 Session 5
9:00 to 10:00 a.m. Thursday, October 21, 2010
Applications of Discrete-Time Competing Risk Hazard Models for Consumer Lending Industry
An accurate and dynamic assessment of profitability (risk & returns) that is sensitive to external market conditions is crucial for the success of any lending business. In particular, in the case of a consumer loan business, a reasonably accurate and dynamic assessment of a customer's future payment behavior is necessary to price a loan product optimally and to plan corrective loss mitigation initiatives. This necessitates estimation of the life of the loan, which in turn requires predicting timing of the loan experiencing default and prepayment events. Though there are a good number of robust statistical methods in use for estimating life of a consumer loan, many times institutions do not prefer these approaches due to greater risk of errors and a larger complexity. This presentation discusses application of discrete-time competing risk hazard models for estimation of the life of a consumer loan by considering product, customer, and market conditions.
The following topics will be covered:
- A brief introduction to the consumer lending business and industry loan product life estimation practices.
- A competing risk hazard modeling approach for simultaneous estimation of a loan default and prepayment incident timings.
- Various applications of these models in consumer lending industry. Sharing model simulation results.
- Potential modeling errors and their impacts.
- Sharing best practices across the industry.
Key Words: Hazard, Default, Prepayment, Competing Risks, Loss, Simulation, Parametric Modeling
Balakerthy Punyakoti and Dheeraj Awasthy
HSBC Bangalore