CMStatistics 2020: Start Registration
View Submission - CFE
Title: Model risk of volatility models Authors:  Emese Lazar - University of Reading (United Kingdom) [presenting]
Ning Zhang - Shanghai University (China)
Abstract: To evaluate the accuracy of volatility models, we propose a new model risk measure and estimation methodology based on loss functions. The reliability of the proposed estimation has been verified via simulations, and the estimates provide a reasonable fit to the true model risk measure. We undertake an empirical analysis based on several assets, identify the models most affected by model risk, and argue that the accuracy of volatility models can be improved by adjusting variance forecasts for model risk. We find that after crises, the model risk increases especially for badly fitting volatility models.