Forecasting oil futures price volatility with economic policy uncertainty: a CARR-MIDAS model

ABSTRACT

In this paper, we propose a conditional autoregressive range-mixed-data sampling (CARR-MIDAS) model that incorporates economic policy uncertainty (EPU) to predict the crude oil futures price volatility (range). We apply the proposed model to West Texas Intermediate (WTI) oil futures price ranges and four EPU indices, namely the Global EPU, US EPU, China EPU and Russia EPU. Empirical results show that all the four EPU indices have a significantly negative impact on the oil futures price volatility, and the EPU indices are informative for forecasting the oil futures price volatility. Moreover, the China EPU index outperforms the other EPU indices in forecasting the oil futures price volatility.

KEYWORDS:

• Economic policy uncertainty
• oil futures price volatility
• price range
• CARR-MIDAS model
• volatility forecasting

JEL CLASSIFICATION:

• C5
• C32
• G17

Acknowledgments

We would like to thank the Editor and an anonymous referee for the insightful comments and suggestions that greatly improved the paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ For comparison, the estimation results for the standard CARR and CARR-MIDAS without EPU (by setting ${\theta }_2=0$ in Eq. (7(6) $h_{i,t}=(1-\alpha -\beta )+\alpha \frac{R_{i-1,t}}{{\tau }_t}+\beta h_{i-1,t}$(6) )) are also reported in Table 2.

² It should be noted that both the HMSE and HMAE are standard statistical loss functions, which are commonly used in the literature. Recently, however, Degiannakis and Filis (2019) argue that volatility forecasts should be evaluated based on the economic evaluation loss functions (Elliott and Timmermann 2008). It would be useful for future research to assess the forecasting performance of our model by employing the economic evaluation criteria. We thank an anonymous referee for pointing this out.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [71971001]; University Natural Science Research Project of Anhui Province [KJ2019A0659]; Innovative Research Project for Graduates of Anhui University of Finance and Economics [ACYC2020106].