ABSTRACT
This study investigates the effect of financial stress (financial stress index; FSI) on oil market realized volatility using Markov-regime switching MIDAS models. The empirical results show that FSI can remarkably improve the forecast accuracy of the oil market. Moreover, FSI measured by the United States can obtain more useful information than that measured by the world. Our findings are robust and reliable via alternative evaluation methods from both statistical and economic aspects.
Acknowledgments
The authors are grateful to the National Natural Science Foundation of PR China [71802167, 71902128, 72071162, 72073109], the Humanities and Social Science Fund of the Ministry of Education [17YJC790105, 17XJCZH002], Sichuan Provincial Philosophy and Social Science Planning Project [SC20TJ004], Sichuan Provincial Science and Technology Planning Project [21RKX0637], Soft Science Research Project in Chengdu [2020-RK00-00070-ZF] and the Fundamental Research Funds for the Central Universities [2682020ZT98].
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
2 The results are also robust in other forecasting windows. Due to the length limitation, we omit them in the manuscript.