Oil price risk evaluation using a novel hybrid model based on time-varying long memory

Abstract

The volatility of crude oil price has a great influence on the world economy. In order to measure the crude oil price risk (VaR) and explain the dynamic relationship between investment income and risk in the oil market more clearly, this paper uses a variety of fractional GARCH models to describe typical volatility characteristics like long memory, volatility clustering, asymmetry and thick tail. The autoregressive conditional heteroscedasticity in the mean model (ARCH-M) and peaks-over-threshold model of extreme value theory (EVT-POT) are taken into account to develop a hybrid time-varying long memory GARCH-M-EVT model for calculation of static and dynamic VaR. Empirical results show that the WTI crude oil has a significantly long memory feature. All the fractional integration GARCH models can describe the long memory appropriately and the FIAPARCH model is the best in regression and out of sample one-step-ahead VaR forecasting. Back-testing results show that the FIAPARCH-M-EVT model is superior to other GARCH-type models which only consider oil price fluctuation characteristics partially and traditional methods including Variance-Covariance and Monte Carlo in price risk measurement. Our conclusions confirm that considering long memory, asymmetry and fat tails in the behavior of energy commodity return combined with effectively dynamic time-varying risk reflection such as the ARCH-M model and reliable tail extreme filter processes such as EVT can improve the accuracy of crude oil price risk measurement, provide an effective tool for analyzing the extreme risk of the tail of the oil market and facilitate the risk management for oil market investors.