Zusammenfassung
Die statistische Qualität von Kreditausfallprognosen lässt sich auf verschiedene Weise messen und vergleichen. Der vorliegende Übersichtsartikel fasst die in der Literatur gemachten Vorschläge zusammen und diskutiert deren Eignung für Kreditausfallprognosen im Privatkundengeschäft. Es zeigt sich, dass nicht alle Qualitätskriterien hier gleichermaßen sinnvoll sind. Insbesondere scheinen die in der Meteorologie beliebten Brier Scores und verwandte Kriterien für diese Anwendungen eher schlecht geeignet.
Abstract
The statistical quality of credit default forecasts can be measured and compared in different ways. This article surveys the various approaches that have been suggested in the literature and discusses their respective properties. For the particular case of credit scoring in the retail business, it is shown that some quality criteria are more useful than others. In particular, various measures that are popular in, e.g. meteorology, such as the Brier score have to be applied with caution.
Literatur
Blöchlinger A, Leippold M (2006) Economic benefit of powerful credit scoring. J Bank Finance 30:851–873
Brier GW (1950) Verification of forecasts expressed in terms of probability. Mon Weather Rev 78:1–3
DeGroot M, Fienberg SE (1983) The comparison and evaluation of probability forecasters. Statistician 32:12–23
Engelmann B, Hayden E, Tasche D (2003) Testing rating accuracy. Risk 16:82–86
Friedman TL (1996): Interview in der “News Hour with Jim Lehrer”, PBS-Televison, 13. Februar
Good IJ (1952) Rational decisions. J R Stat Soc 14:107–114
Hand DJ (2005) Good practice in retail credit scorecard assessment. J Oper Res Soc 56:1109–1117
Hanley A, McNeil BJ (1982) The meaning and use of the area under a receiver operating characteristics (ROC) curve. Diagn Radiol 143:29–36
Hartung J (1987) Statistik, 6. Aufl. Oldenbourg, München
Hoadley B, Oliver RM (1998) Business measures of scorecard benefit. IMA J Math Appl Bus Ind 9:55–64
Krämer W (1998) Measurement of inequality. In: Ullah A, Giles D (Hrsg) Handbook of applied economic statistics. CRC Press, New York, S 39–62
Krämer W (2003) Die Bewertung, der Vergleich von Kreditausfallprognosen. Kredit Kap. 36:395–410
Krämer W (2005) On the ordering of probability forecasts. Sankhya – Ind J Stat 67:662–669
Krämer W (2006) Evaluating probability forecasts in terms of refinement and strictly proper scoring rules. J Forecast 25:223–226
Krämer W, Güttler A (2008) On comparing the accuracy of default predictions in the rating industry. Empir Econ 34:343–356
Krzanowski WJ, Hand DJ (2009) ROC curves for continuous data. Chapman & Hall, Boca Raton
Menkhoff L, Schmeling M, Schmidt U (2010) Are all professional investors sophisticated? Ger Econ Rev 11(3):424–446
Mertens P, Rässler S (2010) Prognoserechnung, 6. Aufl. Physica-Verlag, Heidelberg
Murphy AH (1973) A new vector partition of the probability score. J. Appl. Meteorol. 12:595–600
Rosenberg E, Gleit A (1994) Quantitative methods in credit management: a survey. Oper Res 42:589–613
Sanders F (1963) On subjective probability forecasting. J. Appl. Meteorol. 2:191–201
Stein RM (2005) The relationship between default prediction and lending profits: integrating ROC analysis and loan pricing. J Bank Finance 29:1213–1236
Vardeman S, Meeden G (1983) Calibration, sufficiency and domination considerations for Bayesian probability assessors. J Am Stat Assoc 78:808–816
Winkler RL (1996) Scoring rules and the evaluation of probabilities. Test 5:1–60
Author information
Authors and Affiliations
Corresponding author
Additional information
Diese Arbeit entstand im Rahmen des DFG-Sonderforschungsbereiches 823: „Statistik nichtlinearer dynamischer Prozesse in Wirtschaft und Technik“. Wir danken Karla Schiller, Corinna Großmann, Ulrich Anders, dem Herausgeber Wolfgang Brachinger und einem anonymen Gutachter für zahlreiche Verbesserungsvorschläge und konstruktive Kritik.
Rights and permissions
About this article
Cite this article
Krämer, W., Bücker, M. Probleme des Qualitätsvergleichs von Kreditausfallprognosen. AStA Wirtsch Sozialstat Arch 5, 39–58 (2011). https://doi.org/10.1007/s11943-011-0096-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11943-011-0096-0