Project Repository for QF604.
Team members:
- Caden Lee
- Gourav Kumar Agarwal
- Kenn Ong
The purpose of this study is to replicate and evaluate the results in a study by Guo, Liu, Sun, Zhang, and Ji (2023), titled "Predicting natural gas futures’ volatility using climate risks". In this study, Guo et al. (2023) attempted to model the conditional volatility of the natural gas futures tracking the Henry Hub prices using the GARCH-MIDAS model. Apart from using realized variance as the lower frequency variable – as introduced by Ghysels et al. (2006), Guo et al. included the CPU index and the monthly global climate disaster frequency as additional variables.
As part of this study, the analysis conducted by Guo et al. will be replicated as far as possible to evaluate its performance given new observations. To extend on the study, other factors are to be considered to improve the model performance. Natural gas, like other physical commodities, face demand and supply constraints. Therefore, there is likely to be a strong relationship between price and return volatility with demand and supply factors. Fundamental factors shall be introduced in this extension.