The effect of green energy, global environmental indexes, and stock markets in predicting oil price crashes: Evidence from explainable machine learning

Ben Jabeur, Sami; Khalfaoui, Rabeh; Ben Arfi, Wissal

This study aims to predict oil prices during the 2019 novel coronavirus (COVID-19) pandemic by looking into green energy resources, global environmental indexes (ESG), and stock markets. The study employs advanced machine learning, such as the LightGBM, CatBoost, XGBoost, Random Forest (RF), and neural network models. An accurate forecasting framework can effectively capture the trend of the changes in oil prices and reduce the impact of the COVID-19 pandemic on such prices. Additionally, a large dataset with different asset classes was used to investigate the crash period. The research also introduced SHapely Additive exPlanations (SHAP) values for model analysis and interpretability. The empirical results indicate the superiority of the RF and LightGBM over traditional models. Moreover, this new framework provides favorable explanations of the model performance using the efficient SHAP algorithm. It also highlights the core features of predicting oil prices. The study found that high values of GER and ESG lead to lower crude oil prices. Our results are crucial for investors and policymakers in promoting climate change mitigation and sustained economic prosperity through green energy resources.