Die Trennschärfe von Ratingfunktionen

Abstract

The estimate of creditworthiness represents a central success factor in the loan business of each bank. In particular, it is provided by rating models as a consequence from Basel II. Such models use rating functions to aggregate various information about the rating applicant. The rating allows a statement about the default probability of a debtor.
We discuss different measures of the discriminative power of rating function to examine the forecast strength of ratings. Beside the measures discussed in the literature, we also consult stochastic dominance criteria to evaluate and compare rating functions. The represented characteristics and relations are applied empirically to the ratings of Standard & Poor's and Moody's Investors Service.