Journal
JOURNAL OF EMPIRICAL FINANCE
Volume 52, Issue -, Pages 40-55Publisher
ELSEVIER SCIENCE BV
DOI: 10.1016/j.jempfin.2019.01.004
Keywords
Oil market; Volatility forecasting; Jump intensity; Signed jumps; Cojumps
Categories
Funding
- Natural Science Foundation of China [71671145, 71701170]
- Humanities and Social Science Fund of the Ministry of Education [17YJC790105, 17XJCZH002]
- Fundamental Research Funds for the Central Universities [2682017WCX01, 2682018WXTD05]
Ask authors/readers for more resources
Oil markets are subject to extreme shocks (e.g. Iraq's invasion of Kuwait), causing the oil market price exhibits extreme movements, called jumps (or spikes). These jumps pose challenges on oil market volatility forecasting using conventional volatility dynamic models (e.g. GARCH model) This paper characterizes dynamics of jumps in oil market price using high frequency data from three perspectives: the probability (or intensity) of jump occurrence, the sign (e.g. positive or negative) of jumps, and the concurrence with stock market jumps. And then, the paper exploits predictive ability of these jump-related information for oil market volatility forecasting under the mixed data sampling (MIDAS) modeling framework. Our empirical results show that augmenting standard MIDAS model using the three jump-related information significantly improves the accuracy of oil market volatility forecasting. The jump intensity and negative jump size are particularly useful for predicting future oil volatility. These results are widely consistent across a variety of robustness tests. This work provides new insights on how to forecast oil market volatility in the presence of extreme shocks.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available