Presenting a model for Multiple-step-ahead-Forecasting of volatility and Conditional Value at Risk in fossil energy markets

Fossil energy markets have always been known as strategic and important markets. They have a significant impact on the macro economy and financial markets of the world. The nature of these markets are accompanied by sudden shocks and volatility in the prices. Therefore, they must be controlled and forecasted by using appropriate tools. This paper adopts the Generalized Auto Regressive Conditional Heteroskedasticity (GARCH)-type models, Exponential Smoothing (ES)-type models, and classic model in order to multiple-step-ahead forecast volatility, Value at Risk, and Conditional Value at Risk of Brent oil and natural gas in two different estimation window lengths, respectively. To evaluate the accuracy of the aforementioned models, eight different loss functions are utilized. The results show that, across all forecasting horizons and subsamples used, the Holt-Winters Exponential Smoothing (HWES) model, in comparison with GARCH-type models and classic model, provides more accurate forecasting of the volatility, Value at Risk, and Conditional Value at Risk, respectively. Therefore, the HWES model is proposed to multiple-step-ahead forecast these measures in fossil energy markets.