The situation of a limited availability of historical data is frequently encountered in portfolio risk estimation …, especially in credit risk estimation. This makes it, for example, difficult to find temporal structures with statistical … into account. The modelling framework is based on multivariate elliptical processes which model portfolio risk via sub …

. For this purpose we examine the risk profiles of Collateralized Debt Obligations (CDOs) in some detail. The analyses … reveal significant differences in the risk profile between CDO tranches and corporate bonds, in particular concerning the … considerably increased sensitivity to systematic risks. This has farreaching consequences for risk management, pricing and …

) abilities and on the sectoral concentration risk of a credit portfolio. In this paper, we examine in the first part if … concentrations on the credit risk of the portfolio. Our empirical results suggest that specialization benefits overcompensate the … increase of portfolio risk due to the higher sectoral concentration. If specialization is instead measured by distance measures …

evidence: Rating changes occur relatively seldom, exhibit serial dependence, and lag changes in the issuers’ default risk. In …

, we forecast future leverage ratios and include them in the set of default risk drivers. The analysis is done with a …

Results from portfolio models for credit risk tell us that loan concentration in certain industry sectors can … substantially increase the value-at-risk (VaR). The purpose of this paper is to analyze whether a tractable "infection model" can … provide a meaningful estimate of the impact of concentration risk on the VaR. I apply rather parsimonious data requirements …

This paper discusses duration models for the quantification of credit risk. Econometric techniques to quantify the …

? on credit portfolio risk. Using S&P ratings from 1996 to 2005, we estimate a transition matrix that is insensitive to and … Value-at-Risk (VaR) and the momentum-sensitive VaR. We find realistic scenarios where investors who rely on insensitive … transition matrices underestimate the VaR by eight percent of the correct value. The result is relevant for risk managers and …

which include estimation uncertainty but ignore default correlation might estimate the real credit risk more correctly than …