Interplay of Volatility and Geopolitical Tensions in Clean Energy Markets: A Comprehensive GARCH-LSTM Forecasting Approach

In an era dominated by increasing global challenges and market volatilities, this study, firstly, embarks on an in-depth exploration of volatility transmission across clean energy stocks, crude oil and financial markets, emphasizing the underlying currents of geopolitical tensions. By using the advanced Multivariate Dynamic Conditional Correlation (MV-DCC) GARCH model, we unravel a landscape where volatility spillovers exhibit a distinct bidirectional nature, and geopolitical risk exerts a substantial impact, cascading from the oil market to financial markets and ultimately to clean energy stocks. Our findings underline the strategic importance of overweighting clean energy assets in a dual-asset portfolio that includes oil and financial equities to enhance investment strategies in turbulent market conditions. Secondly, we investigate the predictive power of oil and market-implied volatilities in forecasting clean energy market volatility by introducing a novel approach that melds the robustness of GARCH models with the flexibility of Long Short-Term Memory (LSTM) networks, creating an innovative hybrid GARCH-LSTM framework. The empirical results demonstrate that this hybrid model significantly outstrips the predictive capabilities of traditional standalone models. Notably, while oil and market-implied volatilities substantially enhance prediction accuracy, the inclusion of historical data does not yield additional predictive value. The implications of our research extend beyond the analytical domain, resonating with financial practitioners and environmentally conscious investors who seek precision in valuation and foresight in market trends. For policymakers, the insights provided offer strategic guidance for developing robust clean energy policies. Overall, our research contributes a fresh perspective to the discourse on renewable energy investment, volatility forecasting, and the interplay between market dynamics and geopolitical risks.