Econometric Methods and Monte Carlo Simulations for Financial Risk Management
Abstract: Value-at-Risk (VaR) forecasting in the context of Monte Carlo simulations is evaluated. A range of parametric models is considered, namely the traditional Generalized Autore- gressive Conditional Heteroscedasticity (GARCH) model, the exponential GARCH and the GJR-GARCH, which are put in the context of the Gaussian and Student-t distri- butions. The returns of the S&P 500 provide the basis for the study. Monte Carlo simulations are then applied in the estimation and forecasting of index returns. Two forecasting periods are employed with respect to the Global Financial Crisis (GFC). The forecasting accuracy of the various models will be evaluated in order to determine the applicability of these VaR estimation techniques in dierent market conditions. Results reveal that: (i) no model has consistent performance in both volatile and stable mar- ket conditions; (ii) asymmetric volatility models oer better performance in the post crisis forecasting period; (iii) all models underestimate risk in highly unstable market conditions.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)