Skip to main content
SpringerLink
Log in

What is at stake in the construction and use of credit scores?

  • Published:
Computational Economics Aims and scope Submit manuscript

Abstract

Most of statistical studies on credit scoring focus on scores construction. It is more unusual that they link the statistical technics with a detailed analysis of the users’ requirements regarding the properties of these tools. Concerning companies’ failure the users are financial analysis experts or bankers in credit risk departments or banking supervisors. The increasing need for better control of credit risk by banks has led to a stepping-up of research concerning credit scoring. In the context of the Basel II agreement, the International Banking Committee has stressed the importance of forecasting the expected loss (EL) and, using extreme quantiles, the unexpected loss (UL) for a population of companies, in particular for customers of each commercial bank. In order to do so, it is necessary to estimate the default probability of each company at a given time horizon (PD). The objective of an accurate forecasting gives rise to several needed properties and questions that are presented in Sect. 1. We stress what is at stake in the construction and the use of credit scores. The experience of Banque de France in prudential supervision and the importance of its data files on companies give the possibility to developp a scoring system able to fullfil these needed properties, at least partially. Some principles of credit scoring construction in order to increase the quality of the tool and the accuracy of default probability are presented in Sect. 2. Without leading a complete debate on models’choice we discuss some arguments regarding this choice and we concentrate on comparison between Fisher linear discriminant analysis (LDA) and logistic regression (LOGIT) in Sect. 3. In relation with the early detection of companies default, two pratical uses of a credit scoring system are presented in Sect. 4. Research under way on Banque de France data concentrates on informations that can be extracted from these data on purpose to study how to increase the quality of tools needed by the Basel II agreement. A short overview of this research is given in Sect. 5.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Altman E.I., Saunders A. (1998). Credit risk measurement: Developments over the last 20 years. Journal of Banking and Finance 21: 1721–1742

    Article  Google Scholar 

  • Amemiya T., & Powell J. (1983). A comparison of Logit model and normal discriminant analysis when the independent variables are binary. In Karlin, Amemiya, Goodman, edition Studies in econometrics, time series and multivariate statistics. New York: Academic.

  • Anderson T.W. (1984). An introduction to multivariate statistical analysis. Chapter 6 Classification of observations. New York, Wiley

    Google Scholar 

  • Baesens B., van Gestel T., Viaene S., Stepanova M., Suykens J., & Vanthienen J. (2003). Benchmarking state-of-art classification algorithms for credit scoring. Journal of the Operational Research Society, 9.

  • Bardos M. (2006). Banque de France Scores. Development, applications, and maintenance. Banque de France bulletin Digest no.151 July pp. 1–16

  • Bardos M. (2001). Analyse discriminante: Application au risque et scoring financier. Dunod.

  • Bardos M. (1998a). Le score BDFI: Du diagnostic individuel à l’analyse de portefeuille. Les études de l’Observatoire des entreprises. France, Banque de France

    Google Scholar 

  • Bardos M. (1998b). Detecting the risk of company failure. The Journal of Banking and Finance, 22

  • Bardos M., Foulcher S., Bataille E. (2004). Banque de France scores: method, results, practice. France, Banque de France

    Google Scholar 

  • Bardos M., & Plihon D. (1999). Detection of risk sectors—The IRISK method Banque de France Bulletin Digest no 71, pp. 19–35

  • Bardos M., & Zhu W.H. (1998b). Comparison of linear discriminant analysis and neural networks, application for the detection of company failures. In Bio-mimetic approaches in management science. (pp. 77–100). Dordrecht: Kluwer Academic Publishers.

  • Barret G., & Donald S.G. (2002). Consistent Tests for Stochastic Dominance, working paper, March.

  • BASEL II agrement documentation on BIS website.

  • Breiman L., Freidman J.H., Ohlson R.A., Stone C.J. (1984). Classification and regression trees. California, Edition Wadsworth International Group

    Google Scholar 

  • Bataille E., Bruneau C., Flageollet A., & Michaud F. (2005). Are business cycle and corporate failures in France linked? Computational Economics

  • Blockwitz S., & Hohl S. (2001). Reconciling Ratings, Risk Magazine, June.

  • Celeux G., & Nakache J. P. (1994). Analyse discriminante sur variables qualitatives, Polytechnica.

  • Davidson R., & Jean-Yves D. (1999). Statistical inference for stochastic dominance and for the measurement of poverty and inequality. working paper, November.

  • Efron B. (1975). The efficiency of logistic regression compared to normal discriminant analysis. Journal American Statistical Association 70, 892–898

    Article  Google Scholar 

  • Efron B., Tibshirani R.J. (1993). An introduction to the Bootstrap (ed.). London: Chapman & Hall.

  • Foulcher S., Gouriéroux C., & Tiomo A. (2003). La structure par terme des taux de défauts et rating. Cahiers d’Etudes et Recherches de l’Observatoire des entreprises. France: Banque de France.

  • Foulcher S., Gouriéroux C., & Tiomo A. (2004a). La corrélation de migration: Méthode d’estimation et application aux historiques de notation des entreprises françaises. Cahiers d’ Études et Recherches de l’Observatoire des entreprises. France: Banque de France.

  • Foulcher S., Gouriéroux C., Tiomo A. (2004b). Latent variable approach to dependence of credit risks modelling Cahiers d’ Études et Recherches de l’Observatoire des entreprises. France, Banque de France

    Google Scholar 

  • Gnanadesikan R., et al. (1989). Discriminant analysis and clustering. Statistical Science 4(1): 34–69

    Google Scholar 

  • Gourieroux C. (1989). Économétrie des variables qualitatives. Économica.

  • Hand D.J. (1981). Discrimination and classification. New York, Wiley series in probability and mathematical statistics

    Google Scholar 

  • Hristache M., Delecroix M., & Patiléa V. (2004). On semi parametric m-estimation. Journal of Statistic Planning and Inference.

  • Kendall M.G., & Stuart A. (1961). The advanced Theory of Statistics. Chapter 32, Some Use of Order Statistics, Distribution free, Confidence Intervals for Quantiles. London: Griffin.

  • Mc Lachlan G.J. (1992). Discriminant Analysis and Statistical Pattern Recognition. New-York, Wiley

    Google Scholar 

  • Lebart L., Morineau A., & Piron M. (1998). Statistique Exploratoire Multidimensionnelle, Dunod.

  • Lelogeais L. (2003). Un score sur variables qualitatives pour la détection précoce des défaillances d’entreprises. Bulletin de la. Banque de France 114, 29–46

    Google Scholar 

  • Lo A.W. (1986). Logit versus discriminant analysis. Journal of Econometrics 31, 151–178

    Article  Google Scholar 

  • Maddala G.S. (1999). Limited-dependent and qualitative variables in econometrics. Cambridge, Cambridge University Press

    Google Scholar 

  • Pomaré, T. (2006). Concentration du risque, a student working paper, Observatory of companies.

  • Saporta G. (1990). Probabilités, analyse des données et statistique. Technip.

  • Stili D. (2003). Les incidents de paiement sur effets de commerce. Cahiers d’Etudes et Recherches de l’Observatoire des entreprises. Banque de France.

  • Stili D. (2005). Détection précoce des défaillances d’entreprises et contagion du risque. Thèse de doctorat, Université Paris I.

  • Stili D., & Bardos M. (2006). Risk contagion through defaults in trade bills. Banque de France Bulletin Digest no. 155, November, pp. 1–23

  • Thiria S., Lechevallier Y., Gascuel O., & Canu S. (1997). Statistique et méthodes neuronales. Dunod.

  • Tiomo A. (2002). Risque de crédit et variabilité des taux de défaut: une analyse empirique par simulation, Etudes et Recherches de l’Observatoire des entreprises. France, Banque de France

    Google Scholar 

  • Vessereau A. (1987). Une propriété peu connue: l’intervalle de confiance de la médiane. Revue de Statistique Appliquée XXXV(1): 5–8

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Companies Observatory, Banque de France, Paris, France

    Mireille Bardos

Authors
  1. Mireille Bardos
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Mireille Bardos.

Additional information

“Statistical inference techniques, if not applied to the real world, will lose their import and appear to be deductive exercises. Furthermore, it is my belief that a statistical course emphasis should be given to both mathematical theory of statistics and to application of the theory to practical problems. A detailed discussion on the application of a statistical technique facilitates better understanding of the theory behind the technique.” C. Radhakrishna RAO in Linear Statistical Inference and Its Applications

Rights and permissions

Reprints and permissions

About this article

Cite this article

Bardos, M. What is at stake in the construction and use of credit scores?. Comput Econ 29, 159–172 (2007). https://doi.org/10.1007/s10614-006-9083-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10614-006-9083-x

Keywords

Search

Navigation