Modeling value-at-risk for green bonds and clean energy investments

This study employs 40 GARCH-based VaR models, assuming Gaussian and non-Gaussian distributional innovations to model the risk in green bonds and clean energy investments, addressing a critical gap in existing literature. By utilizing the Model Confidence Set (MCS) technique to generate superior set models (SSMs) and rank them based on the predictive performance of Value-at-Risk (VaR) forecasts, this research offers a more robust and rigorous approach for risk model selection. We employ 1601 daily log-returns of green bond and clean energy, which span 03/12/2017 to 13/12/2023. We find that for both 1% and 5% VaR forecasts, green bonds demonstrate greater heterogeneity across the models compared to clean energy. Specifically, green bonds have the fewest models included in the SSM. Our findings offer valuable insights into the unique risk dynamics of sustainable finance, risk modeling, and risk management. In contributing to the stability of financial systems as they adapt to the global transition toward a low-carbon and sustainable economy, this could not be more important.