Joint Statistics and Computational Finance Seminar
Given the growing need for managing and reporting financial risk, risk prediction plays an increasing role in banking and finance. In this study, we compare the out-of-sample performance of existing methods and some new models for predicting Value-at-Risk in a univariate context. Using more than 30 years of daily return data on the NASDAQ Composite Index, we find that most approaches perform inadequately, although several models are acceptable under current regulatory assessment rules for model adequacy. A hybrid method, combining a heavy-tailed GARCH filter with an extreme value theory-based approach, performs best overall, closely followed by a variant on a filtered historical simulation, and a new model based on heteroscedastic mixture distributions. Conditional autoregressive VaR (CaViaR) models perform inadequately, though an extension to a particular CaViaR model is shown to outperform the others.
* Stefan Mittnik is Chair of Financial Econometrics, Institute of Statistics, University of Munich, and Program Director of Financial Risk Management at the Center for Financial Studies in Frankfurt. During the current academic year he is a Fulbright Scholar at Washington University in St. Louis, where he received the Ph.D. in Economics in 1987. He is the author of many publications, including the definitive book Stable Paretian Models in Finance co-authored with S. T. Rachev (2000, Wiley).