Estimating Probabilities of Default With Support Vector Machines

DefaultRisk.com the web's biggest credit risk modeling resource.

Submit Your Paper

Post Your Résumé

For Recruiters

Today's Featured Book

Applied Quantitative Finance
by Wolfgang K. Härdle (Editor), Nikolaus Hautsch (Editor), Ludger Overbeck (Editor), Springer, September 1, 2008, Hardcover, 448 pages

Training Discounted for DefaultRisk.com visitors only:

The Mathematics of Credit Derivatives: The Essential Credit Modelling and Pricing Companion
by Philipp J. Schönbucher,
WBS Training, August 2003, DVD / Interactive CD-ROM

Sponsor:
Shop at Amazon.com and support DefaultRisk.com

Estimating Probabilities of Default With Support Vector Machines

by Wolfgang K. Härdle of Humboldt-Universität zu Berlin,
Rouslan A. Moro of the German Institute for Economic Research & Humboldt-Universität zu Berlin, and
Dorothea Schäfer of the German Institute for Economic Research

May 27, 2007

Abstract: This paper proposes a rating methodology that is based on a non-linear classification method, the support vector machine, and a non-parametric technique for mapping rating scores into probabilities of default. We give an introduction to underlying statistical models and represent the results of testing our approach on German Bundesbank data. In particular we discuss the selection of variables and give a comparison with more traditional approaches such as discriminant analysis and the logit regression. The results demonstrate that the SVM has clear advantages over these methods for all variables tested.

JEL Classification: C14, G33, C45.

Keywords: Bankruptcy, Company rating, Default probability, Support vector machines.

Books Referenced in this Paper: (what is this?)