True data-generating process of energy volatility dynamics

Determine the true data-generating process that governs the dynamics of volatility in major energy commodity prices (such as crude oil, natural gas, and gasoline) to provide a validated foundation for forecasting and risk management models.

Background

The paper motivates the importance of accurately modeling and forecasting the volatility of energy commodities due to their economic impact and the role of volatility in pricing, risk management, and macroeconomic analysis. Despite numerous models (including those incorporating structural breaks, regime switching, heterogeneous autoregressive structures, and long memory), many approaches rely on assumptions that are difficult to validate.

The authors propose a Time-Varying Extended Wold Decomposition (TV-EWD) to identify and model heterogeneous, time-varying persistence in volatility shocks. However, they explicitly acknowledge that identifying the true data-generating process underlying volatility dynamics remains unresolved, highlighting a fundamental open question in the field.