Backtesting and Benchmarking Credit Risk Models beyond 2008: a SOA based information system architecture.

The introduction of compliance guidelines such as Basel II, Sarbanes-Oxley, and Solvency II has a huge impact on the investments and strategies of financial institutions nowadays. The Basel II Capital Accord aims at quantifying the minimum amount of regulatory buffer capital so as to provide a safety cushion against unexpected credit-, market- and/or operational losses (Basel Committee on Banking Supervision, 2006). From a credit-risk perspective, the Accord encourages financial institutions to build risk models hereby using three key risk-parameters: Probability of Default (PD), Loss Given Default (LGD), and Exposure at Default (EAD). The research team of prof. Baesens and prof. Vanthienen has already extensively investigated how these models can be built using data mining techniques such as neural networks, decision trees, rule extraction techniques, business rules and decision tables. Results of this research have been published in well-respected journals and/or presented at international conferences (Baesens et al. 2002, Baesens et al 2003a, Baesens et al. 2003b, Baesens et al. 2006, Martens et al. 2007). With respect to service-orientation, Prof. Poelmans has conducted research on service granularity and its architectural impact (Haesen et al. 2008). By developing and validating IS usability evaluation models, prof. Poelmans is also acquainted with advanced quantitative validation tasks (Poelmans et al. 2008, Reijers et al. 2007). The purpose of this project is to investigate and study the validation of the estimated PD, LGD and EAD risk models. Validating Basel II risk models is a very challenging and complex activity whereby a distinction can be made between quantitative and qualitative validation. Quantitative validation aims at using quantitative techniques (data mining, statistics, ) combined with data. The most important qualitative validation tasks to be performed are setting up a data quality program, conducting a use test (to which extent are the Basel II risk models consistently used throughout the organization), and the involvement of senior management in the Basel II implementation and validation process. In this project, the major focus will be on quantitative validation.

Keywords:SOA based information system architectur

Disciplines:Applied economics