Forecasting oil and gold volatilities with sentiment indicators under structural breaks

This paper contributes to the literature on forecasting realized volatility of oil and gold by (i) utilizing Infinite Hidden Markov (IHM) switching model within Heterogeneous Autoregressive (HAR) framework accommodate structural breaks in data (ii) incorporating, for first time literature, various sentiment indicators that proxy speculative hedging tendencies investors these markets as predictors models. We show accounting incorporating sentiment-related does not only improve out-of-sample performance models but also has significant economic implications, offering improved risk-adjusted returns investors, particularly short-term mid-term forecasts. find evidence cross-market information spilling over across oil, gold, stock predictability market fluctuations due factors. The results highlight predictive role investor factors improving forecast accuracy dynamics commodities with potential yield gains markets.