A1397
Title: Estimation risk for systemic risk measures driven by semi-parametric models
Authors: Jeremy Leymarie - ESC Clermont Business School (France) [presenting]
Abstract: The purpose is to provide an analytical method to quantify the estimation risk contained in the systemic risk measures used to identify the financial institutions that contribute the most to the overall risk in the financial system. Estimation of the marginal expected shortfall (MES) and the delta conditional value-at-risk (Delta CoVaR) are investigated, when the firm and market returns follow a semi-parametric bivariate dynamic model. A two-step filtered historical simulation method, which is assumption-free on the distribution of the innovations, is proposed to estimate these quantities. We develop the asymptotic theory for the MES and the Delta CoVaR. The proposed method is evaluated by simulation and proved valid. Then, the method is applied to a panel of U.S. financial institutions to test whether the systemic risk contribution of a financial entity is significant on a single day, e.g. on September 15th 2008, given the information set prior to this date.
