Risk Measure Inference

Mardi | 2015-03-24
Sully 5, 16h-17h20

Christophe HURLIN – Sébastien LAURENT – Rogier QUAEDVLIEG – Stephan SMEEKES

We propose a widely applicable bootstrap based test of the null hypothesis of equality of two fi rms’ Risk Measures (RMs) at a single point in time. The test can be applied to any market-based measure. In an iterative procedure, we can identify a complete grouped ranking of the RMs, with particular application to fi nding buckets of fi rms of equal systemic risk. An extensive Monte Carlo Simulation shows desirable properties. We provide an application on a sample of 94 U.S. financial institutions using the CoVaR, MES and %SRISK, and conclude only the %SRISK can be estimated with enough precision to allow for a meaningful ranking.