« Leaving Asilomar | Main | Money words »

Validating Default Probabilities on Short Time Series

Two approaches to examine the accuracy of default probability forecasts
for different rating grades and in particular, the respective
advantages and disadvantages of the two methods. Also, the effect of
independence assumptions is taken into account by modelling latent
variables like the asset correlation and dependency in time. Both
tests, the Extended Traffic Light Approach as well as an ad hoc normal
test work on time-varying default probability forecasts. They are
considered with respect to their practical use and potential
application in validating default forecasts in credit institutions.

Keywords:
Basel II, Internal Ratings Based Approach, Validation, Default Probabilities

Stefan Blochwitz of Deutsche Bundesbank,
Stefan Hohl of the Bank for International Settlements,
Dirk Tasche of Deutsche Bundesbank, and
Carsten When of Deutsche Bundesbank

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)