ESG stock markets and clean energy prices prediction: Insights from advanced machine learning

Ghallabi, Fahmi; Souissi, Bilel; Du, Anna Min; Ali, Shoaib

Abstract

In the post-Paris agreement, the clean energy market has grown significantly due to its undeniable environmental sustainability. Therefore, this study aims to predict clean energy stock prices by analyzing ESG stock markets in ten countries using a batter of machine learning (ML) techniques and NGBoost. The analysis integrates Shapley Additive Explanations (SHAP) values to improve interpretability, offering insights into model performance. The dataset spans from January 1, 2014, to September 22, 2023, covering global crises such as the COVID-19 pandemic and the Russia-Ukraine conflict. Results indicate that NGBoost outperforms other models, with a significant correlation between clean energy stock prices and ESG market variables. Notably, ESG markets in India and the USA show strong predictive power for clean energy stocks, while those in Australia and South Africa contribute less. These findings underscore the potential of ML techniques in forecasting clean energy equity trends, providing insights for investors, policymakers, and venture capitalists. The study highlights the importance of considering the degree of market connectivity in portfolio construction, emphasizing a shift from traditional investments to sustainable ones like clean energy. This research adds value to clean energy market analysis by incorporating advanced ML methods and SHAP values, especially during periods of global disruption. These results are important for asset allocation and risk management, supporting investors in transitioning from ordinary to sustainable investments.

Citation

Ghallabi, F., Souissi, B., Du, A. M., & Ali, S. (2025). ESG stock markets and clean energy prices prediction: Insights from advanced machine learning. International Review of Financial Analysis, Article 103889. https://doi.org/10.1016/j.irfa.2024.103889

• Journal Article Type: Article
• Acceptance Date: Dec 15, 2024
• Online Publication Date: Dec 17, 2024
• Publication Date: 2025-01
• Deposit Date: Dec 20, 2024
• Publicly Available Date: Dec 20, 2024
• Journal: International Review of Financial Analysis
• Print ISSN: 1057-5219
• Electronic ISSN: 1873-8079
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Article Number: 103889
• DOI: https://doi.org/10.1016/j.irfa.2024.103889
• Keywords: Clean energy; ESG; Machine learning methods