Performance comparison of alternative stochastic volatility models and its determinants in energy futures: COVID‐19 and Russia–Ukraine conflict features
比较了七种随机波动模型在原油、天然气和汽油期货上的表现，发现t分布创新模型最优，且COVID-19和俄乌冲突期间波动特征不同，对短期预测有不同模型偏好。
Abstract This paper studies the volatility dynamics of futures contracts on crude oil, natural gas, and gasoline. An appropriate Bayesian model comparison exercise between seven stochastic volatility (SV) models is estimated using daily prices for our futures contracts between 2005 and 2023. Moreover, to assess the impacts of COVID‐19 and the Russia–Ukraine conflict on volatility, we analyze these two subsamples. Overall, we find that: (i) the Bayes factor shows that the SV model with ‐distributed innovations outperforms the competing models; (ii) crude oil contracts with different expiry dates may require the introduction of leverage effects; (iii) the ‐distributed innovations remain the appropriate model for the COVID‐19 subsample, while jumps are needed in the conflict period; and (iv) other Bayesian criteria more appropriate to short‐term predictive ability—such as the conditional and the observed‐date deviance information criterion—suggest other rank order to model our futures contracts, despite the agreements for the best models.