An Internal Rating Modeling Framework for Low-Default Portfolios

A core input to credit risk measurement and the development of portfolio strategies is the Probability of Default (PD) per borrower. As such, any methodology that derives PD estimates will determine the accuracy and soundness of the results derived from the models, as well as the quality of the decisions that are made using the PD estimates. In this paper, a framework for dealing with low-default portfolios is presented. This framework is based on a shadow rating approach. While the specific project in this paper was performed on large corporate entities, this approach can be applied to both SME and consumer sectors where external ratings are available on at least part of the portfolio.