Title: Scoring function-based model risk of risk models
Authors: Ning Zhang - ICMA Center, University of Reading (United Kingdom)
Emese Lazar - University of Reading (United Kingdom) [presenting]
Radu Tunaru - University of Kent (United Kingdom)
Abstract: Studying the accuracy of risk measures, we propose a scoring function-based model risk estimation methodology, which opens the possibility of measuring joint model risk of the pair of risk measures (VaR, ES) at significance level. We carry out a simulation study to illustrate and analyze our proposed model risk measure across various (VaR, ES) risk models, and also study the properties of the proposed methodology as a measure of (model) risk. An empirical analysis illustrates its application for different asset classes.