Predicting natural gas futures’ volatility using climate risks

Kun Guo,Fengqi Liu,Xiaolei Sun,Dayong Zhang,Qiang Ji
DOI: https://doi.org/10.1016/j.frl.2023.103915
IF: 9.848
2023-04-01
Finance Research Letters
Abstract:In this paper, we examine the tracking and predictive power of two kinds of climate risks—namely, climate policy uncertainty (CPU) and climate-related disasters—on the price volatility of natural gas futures. The GARCH-MIDAS model was adopted to incorporate daily natural gas futures prices with monthly CPU indices and disaster frequencies. The empirical results showed a robust predictive relationship between disaster frequency and natural gas price volatility under both in-sample and out-of-sample scenarios, while combining the CPU index with other predictors could not improve the out-of-sample forecasting performance. We believe these findings could provide insights for traders and market regulators.
business, finance
What problem does this paper attempt to address?