Forecasting volatilities of oil and gas assets: A comparison of GAS, GARCH, and EGARCH models

Abstract

This paper compares Generalized Autoregressive Score (GAS) models and GARCH-type models on their forecasting abilities for crude oil and natural gas spot and futures returns from developing and developed markets over multiple horizons. The out-of-sample forecasting results based on two loss functions and the Diebold-Mariano predictive accuracy test for multiple models show that the GAS framework outperforms GARCH and EGARCH models, particularly for crude oil assets. For natural gas, no specific model retains an advantage

The effectiveness of crude oil futures hedging during infectious disease outbreaks in the 21st century

“…This is especially true when modeling the price dynamics of crude oil. Consequently, a narrow focus on notable unanticipated events, such as those related to pandemics, is required to capture the nonnegligible impacts using GARCH estimates (Xu & Lien, 2022).…”

Bayesian Predictive Distributions of Oil Returns Using Mixed Data Sampling Volatility Models

“…The authors find that for out-of-sample forecasting, fat-tailed GARCH models with leverage effect are hard to beat. Xu and Lien (2022) employ the GAS model with student errors for multistep ahead oil volatility forecasting. The authors find that the GAS model overwhelmingly beats GARCH and EGARCH specifications for 1, 5, and 20-step ahead, but not 60-day ahead forecasts.…”

A hybrid deep learning approach by integrating extreme gradient boosting‐long short‐term memory with generalized autoregressive conditional heteroscedasticity family models for natural gas load volatility prediction

“…The EGARCH model is also an asymmetric model, 35 but the difference is that the variance equation analyzes not $${\sigma }_{t}^{2}$$ but $$\mathrm{ln}\,{\sigma }_{t}^{2}$$. When the estimated value of $$\beta \,\mathrm{ln}\,{\sigma }_{t-1}^{2}$$ in the variance equation is a positive number, it indicates that the sequence has a persistent phenomenon of fluctuations (cluster phenomenon).…”