|
Abstract:
|
The Basel Committee on Banking Supervision published a new proposal (Basel II) for revising the existing Basel I capital adequacy framework from 1999. Basel II forces banks to develop credit risk models to be used in risk management and for calculation of reserves. According to the Basel II guidelines, these models have to be validated. However, regulators have not provided clear guidelines as to how this validation should take place. We propose a simple validation methodology that can be used by banks to validate their credit risk models and apply this methodology to a real-world case. We focus on statistical criteria for validation. Validation is obviously not only a statistical exercise, we also pay attention to managerial judgment and a qualitative analysis of the model. This paper gives a clear overview of the available statistical validation methods.
|
