A note on oil price shocks and the forecastability of gold realized volatility

ABSTRACT

We examine the predictive power of disentangled oil price shocks over gold market volatility via the heterogeneous autoregressive realized volatility (HAR-RV) model. Our in- and out-of-sample tests show that combining the information from both oil supply and demand shocks with the innovations associated with financial market risks improves the forecast accuracy of realized volatility of gold. While financial risk shocks are important on their own, including oil price shocks in the model provides additional forecasting power in out-of-sample tests. Compared to the benchmark HAR-RV model, the extended model with all the three shocks included outperforms, in a statistically significant manner, all other variants of the HAR-RV framework for short-, medium, and long-run forecasting horizons. The findings highlight the predictive power of cross-market information in commodities and suggest that dise ntangling supply- and demand-related factors associated with price shocks could help improve the accuracy of forecasting models.

KEYWORDS:

• Oil shocks
• risk Shocks
• gold
• realized volatility
• forecasting

JEL CLASSIFICATION:

• C22
• C53
• Q02

Acknowledgments

We would like to thank an anonymous referee for many helpful comments. However, any remaining errors are solely ours.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ See Reboredo (2013), Balcilar, Ozdemir, and Shahbaz (2019) and Tiwari et al. (2019) for detailed literature reviews.

² These data are all derived from the Datastream database as maintained by Thomson Reuters. The world integrated oil and gas producer index represents the stock prices of global oil producer companies and includes large publicly traded oil-producing firms (i.e. BP, Chevron, Exxon, Petrobras or Repsol), but not nationalized oil producers (such as ADNOC or Saudi Aramco).

³ In a sense, one can argue that supply shocks in this framework relate to region-specific or event-specific information that cannot be accounted for by stock-market-related pricing effects.

⁴ The discussion on the break dates that follows relies on the discussions in Baumeister and Kilian (2016).

⁵ The absolute MSFEs from the benchmark HAR-RV model at h = 1, 5 and 22 are found to be 0.93%, 1.57% and 2.16%, respectively. These values can, in turn, be used to recover the absolute MSFEs of the extended versions of the HAR-RV model by interested readers.

⁶ The relatively weaker performance at h = 5 is probably due to the insignificant coefficient on the demand shock, which we observe in the corresponding in-sample regression.

⁷ The critical values at 10%, 5% and 1% are 0.1270, 1.6120, and 4.1840 respectively, as derived from Table 4 of McCracken (2007, 732).