A Comparative Forecasting Analysis of Clean Energy Stocks using Recurrent Neural Networks

Climate change represents one of the most pressing existential threats of our time, requiring coordinated, cross-domain responses that integrate technological, financial, and policy-oriented knowledge. This paper investigates the behavior of selected clean energy stock indices before, during, and after the COVID-19 crisis and applies advanced machine learning methodologies, specifically Recurrent Neural Networks (RNNs) and Gated Recurrent Units (GRUs), to predict clean energy stock prices. The results provide new insights into the nonlinear dynamics of financial markets linked to the clean energy sector and show that both LSTM and GRU models outperform VAR in stock price forecasting, delivering superior accuracy. This research highlights the effectiveness of integrating traditional statistical models with deep learning techniques to improve forecasting performance. It promotes a deeper understanding of the behavior of this crucial industry, providing a bridge between finance, technology, and sustainability topics, necessary to achieving a resilient and equitable low-carbon economy.