TY - JOUR AU - Jakob, Kevin AU - Fischer, Matthias PY - 2016/02/29 Y2 - 2024/03/29 TI - GCPM: A ?exible package to explore credit portfolio risk JF - Austrian Journal of Statistics JA - AJS VL - 45 IS - 1 SE - Special Issue on R DO - 10.17713/ajs.v45i1.87 UR - https://ajs.or.at/index.php/ajs/article/view/vol45-1-2 SP - 25-44 AB - In this article we introduce the novel GCPM package, which represents a generalized credit portfolio model framework. The package includes two of the most popular mod- eling approaches in the banking industry namely the CreditRisk+ and the CreditMetrics model and allows to perform several sensitivity analysis with respect to distributional or functional assumptions. Therefore, besides the pure quanti?cation of credit portfolio risk, the package can be used to explore certain aspects of model risk individually for every arbitrary credit portfolio. In order to guarantee maximum ?exibility, most of the models utilize a Monte Carlo simulation, which is implemented in C++, to achieve the loss dis- tribution. Furthermore, the package also o?ers the possibilities to apply simple pooling techniques to speed up calculations for large portfolios as well as a general importance sample approach. The article concludes with a comprehensive example demonstrating the ?exibility of the package. ER -