Exploring the influence of the main factors on the crude oil price volatility: An analysis based on GARCH-MIDAS model with Lasso approach

This paper comprehensively explores various influencing factors of crude oil price volatility from four perspectives: commodity attributes, macroeconomics factors, geopolitical events and alternative energy. On this basis, the generalized autoregressive conditional heteroscedasticity mixed frequency data sampling model (GARCH-MIDAS) is constructed with both level effect and volatility effect, single and multi-factor models. For variable selection in multi-factor models, the Lasso-adaptive method is utilized to solve multicollinearity problems. The findings show that the prediction performance of multi-factor models is better than single-factor models. In the long run, supply and demand continue to be the most influential factors of oil price volatility; inventories, the US dollar exchange rate and geopolitical risk all affect oil price volatility to roughly the same extent; alternative energy can have an impact on oil price fluctuations, but this impact is relatively minor.