Ratings are forecasts, i.e. there are no "right" or "wrong" ratings. However, there are more or less sound ratings. As a rating agency, we attach central importance to the quality of the rating models we use. In doing so, we not only orientate ourselves on the common statistical test procedures of model validation, the results of which lose a great deal of significance in segments and sectors with rare defaults. In assessing the quality of our rating models, we also use qualitative assessment criteria. These include the robustness of the rating architecture in relation to the fundamentals and findings of economics (position) or feedback on the mutual assessment of the rated issuers themselves (ranking).
