Current research team

Bart Baesens	Supervisor
Wouter Verbeke	PhD
Bram Decroix	PhD
Elisabeth Van Laere	PhD
Karel Dejaeger	PhD

Description Credit Scoring and Basel II

Given the recent turmoil on credit markets, the topic of credit risk modelling has now become more important than ever before. The introduction of compliance guidelines such as Basel 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. 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 PD quantifies the probability that an obligor will default typically in the upcoming one year. The LGD parameter measures the economic loss, expressed as percentage of the exposure at default (EAD) , in case of default. All three risk parameters then serve as inputs to an economic capital (ECAP) portfolio model that calculates the loss distribution of a portfolio of exposures, and as such determines the minimum capital (shareholder equity, retained earnings, ...) a financial institution should maintain in order to sufficiently protect its savings depositors.

The research team of professor Baesens currently studies the following topics in this context:

• Developing a new and improved holistic and integrated Economic Capital (ECAP) portfolio model. In contrast to the portfolio models currently available, we plan to address risk at a firm-wide level, by studying how to optimally aggregate and harmonize the various risk measurements available, hereby not only focusing on credit risk as such, but also on the interplay with e.g. market risk, interest rate risk, liquidity risk, concentration risk and non-material risks.
• Finding new and better risk drivers for the key ECAP model parameters such as LGD, EAD, and correlations. Up till now, little is known about what actually drives loss and exposure risk. Nevertheless, these parameters are critical for calculating the economic buffer capital, and knowing their risk drivers is essential.
• Investigating new ways of statistically modelling LGD and EAD. Although modelling methodologies for PD are fairly well established, much progress is still to be made in the LGD and EAD area. Both risk parameters are assumed to have bimodal distributions which highly complicates the statistical modelling. Hence, new statistical procedures are needed to develop predictive but also interpretable LGD and EAD models. The optimal balance between model performance and model interpretability is also a key topic of study.
• New statistical ways of calibrating ECAP parameters. Traditionally, credit risk models aim to discriminate obligors and/or exposures in terms of default (PD), loss (LGD) and exposure (EAD) risk. However, losses need to be absorbed by capital in an absolute way! Hence, calibration procedures should be developed that come up with optimal cardinal measures of risk taking into account both past experience and future expectations.
• Developing new approaches for stress testing the ECAP model. Using stress testing, it can be seen how economic capital is affected by extreme, adverse economic circumstances, as we have seen lately happening in the credit markets. New contributions are needed here, in order to appropriately define stress scenarios, integrate various types of risk, understand the behavior in the tail of the loss distribution, and gauge the impact on credit risk and capital positions at the firm level.
• Coming up with new ways of quantifying and dealing with model risk. Risk quantification models are not perfect and as such also have risk associated with them. It is important to be aware of this and appropriately act upon it, using e.g. conservative parameter calibration and/or other innovative approaches.
• Devising innovative approaches for monitoring and validating the ECAP model parameters. We hereby study new approaches for model backtesting, model benchmarking and qualitative validation of the various ECAP parameters involved.
• Developing new simulation models for better quantifying and assessing systemic risk at the macro level using network based learning. Systemic risk is the risk of failing of a complete system or market. This risk arises from the dependencies and network structures between the various market participants (banks, insurance companies, ...).

Past research