Monday, April 5, 2010

The Predictive Accuracy of Credit Ratings: Measurement and Statistical Inference

By Walter Orth

Abstract: Credit ratings are ordinal predictions for the default risk of an obligor. To evaluate the accuracy of such predictions commonly used measures are the Accuracy Ratio or, equivalently, the Area under the ROC curve. The disadvantage of these measures is that they treat default as a binary variable thereby neglecting the timing of the default events and also not using the full information from censored observations. The problem gets more severe as the prediction horizon is extended. We present an alternative measure that is related to the Accuracy Ratio taking into account the characteristics of the survival data structure which we usually find in rating datasets. We also derive methods to perform statistical inference for the Accuracy Ratio and the proposed new measure, which is a nontrivial task in a setting that includes multiple cohorts of obligors. All procedures are also analyzed empirically using a large dataset of S&P Long Term Credit Ratings.

Download here:

No comments: