Calibrating Low-default Portfolios, using the Cumulative Accuracy Profile
by Marco J. van der Burgt of ABN AMRO
Abstract: In the new Basel II Accord, banks are allowed to develop their own credit rating models. However, the lack of sufficient (default) data for backtesting rating models for "low-default portfolios" is a main concern for the financial industry and regulators. These low-default portfolios are characterized by the lack of sufficient data. In this article we present a method of calibrating low-default portfolios, based on modeling the observed power curve and deriving the calibration from this curve. The curve is determined by a concavity parameter, which can easily be related to the accuracy ratio (AR). The method is demonstrated for sovereign ratings.
Keywords: low-default portfolio, power curve, accuracy ratio.
Published in: Journal of Risk Model Validation, Vol. 1, No. 4, (Winter 2007/08).