Finance Center Muenster

The newly-gained attention towards liquidity risk led to increased demand for a sound risk identification of contingent products like credit facilities. In order to reliably assess the liquidity risk of credit facilities on a portfolio level, the eligibility of the applied portfolio measure must be well-founded. However, to quantitatively evaluate the performance of a spe-cific measure, the (potentially unknown) future portfolio draw-down distribution is needed. We address this problem by constructing an environment for the generation of data sets for varying assumptions. Data histories are simulated for a wide parameter range, allowing us to study the performances of different portfolio measures currently discussed in literature, for multiple scenarios. Via a criteria-based analysis we find that a Monte-Carlo simulation approach built around the modeling of risk factors which incorporates an additional portfo-lio correlation structure, provides the best results for the estimation of a future draw-down distribution