Modelling correlations in credit portfolio risk
Abstract
A credit portfolio’s risk level depends on correlations between latent covariates, such as the probability of default in different economic sectors. correlations often have to be estimated from relatively short time series, and the resulting estimation error hinders the detection of a signal. this paper suggests a general method of parameter estimation which avoids, in a controlled way, the underestimation of correlation risk. the paper presents empirical evidence to show how, in the framework of the creditrisk+ model with integrated correlations, this method leads to an increased economic capital estimate. in this way, the limits of detecting the portfolio’s diversification potential are adequately reflected.
The full article is available to subscribers to the journal.
Author's Biography
Bernd Rosenow received a PhD in physics from Würzburg University in 1998 and joined the Physics Department at Cologne University as a research associate in the same year. He obtained an assistant professorship at Cologne University in 2002. Since 2005, he has been a Heisenberg fellow of the German Research Council, conducting theoretical physics research at Harvard University and the Max-Planck Institute Stuttgart. His research focuses on the analysis of correlations in financial markets, multivariate volatility models, and market microstructure. He uses the theory of random matrices and other methods of mathematical physics to study these problems. His work has been published in international journals including Quantitative Finance and the Journal of Economic Dynamics & Control.
Rafael Weissbach has a diploma in mathematics from Göttingen University and a PhD in statistics from the University of Dortmund. In 2004, he joined the University of Dortmund's Faculty of Statistics as a research assistant and was promoted to assistant professor for econometrics in 2007. Rafael has also acted as chair for econometrics at the University of Mannheim's Faculty of Economics. From 2001 to 2004, he worked as a risk analyst and portfolio manager in the credit risk management division of an international investment bank. His current interest is statistics in finance, especially estimation and the modelling of credit risk related parameters such as rating migration matrices and default correlations. His work has been published in international journals including the Journal of the American Statistical Association. He is currently associate professor for statistics at Rostock University.
Citation
Rosenow, Bernd and Weissbach, Rafael (2010, January 1). Modelling correlations in credit portfolio risk. In the Journal of Risk Management in Financial Institutions, Volume 3, Issue 1. https://doi.org/10.69554/QIIN6993.
Publications LLP