Forecasting Value at Risk (VAR) in the futures market using Hybrid method of Neural Networks and GARCH model

This study proposes a hybrid model, which combines GARCH and Neural Network, for estimating VAR in Nasdaq 100 and Dow Jones futures index market. Empirical results demonstrated that the hybrid method has certain outperformed the conventional method (historical simulation, variance/covariance and the Monte Carlo simulation) in estimating VAR. In terms of accuracy, the hybrid method is superior to any of the conventional methods, especially in the Nasdaq 100 futures index market. In terms of conservativeness, the hybrid method was superior to the HS method in both markets, and hybrid method is superior to the conventional methods in the Nasdaq 100 futures index. In terms of efficiency, the hybrid method were more efficient than HS method when applied to in both futures index, and hybrid method is superior to the conventional methods in the Dow Jones futures index.