Dynamic volatility connectedness between clean and fossil energy indexes: evidence from Generalized Autoregressive Score (GAS) Time-Varying Copulas Analysis, Mehdi Mili and Ebrahim Sohrab

Mehdi Mili and Ebrahim Sohrab
College of Business Administration
University of Bahrain
Kingdom of Bahrain

Abstract: This study investigates the evolving interrelationships between clean energy and fossil fuel markets, with a focus on the implications for sustainable energy transition and risk management in a volatile geopolitical and climate-sensitive context. Using a dynamic econometric framework that combines the Glosten-Jagannathan-Runkle GARCH (GJR-GARCH) model with a Generalized Autoregressive Score (GAS) copula, we analyze tail dependence and volatility spillovers between clean energy indices and key fossil energy markets—namely Natural Gas, Brent Crude, and Heating Oil. Our findings reveal nonlinear dependencies and significant tail risk connectedness. These insights are particularly relevant in light of growing global emphasis on renewable energy adoption, climate resilience, and energy security, especially in regions susceptible to geopolitical instability. The superior forecasting performance of the time-varying GAS copula highlights its practical value for sustainable financial strategies and climate-aware investment decisions. Moreover, by incorporating advanced risk measures such as Value at Risk (VaR) and Expected Shortfall (ES), we demonstrate that diversified portfolios containing both clean and fossil energy assets offer improved risk profiles under dynamic market conditions. This research contributes to the broader dialogue on smart governance, climate risk modeling, and sustainable finance, and aligns with the objectives of WASD’s Sustainable Development Goals Universities Initiative (SDGsUNi), particularly in the context of Gulf and Kuwait universities’ engagement with clean technologies and sustainable development policies.

Keywords: Clean Energy, Fossil Energy, Volatility Persistence, Dynamic Correlations, Multivariate GAS Copula, Value at Risk, Expected Shortfall.