Oil volatility risk and stock market volatility predictability: Evidence from G7 countries

• Published in: Energy economics 2017-10, Vol.68, p.240-254
• Main Authors: Feng, Jiabao, Wang, Yudong, Yin, Libo
• Format: Article
• Language: English
• Subjects: Crude oil; Economic forecasting; Economic significance; Energy economics; Forecasting; Low level; Macroeconomics; Oil; Oil volatility risk premium; Out-of-sample forecast; Petroleum; Power; Predictions; Predictive regression; Risk; Securities markets; Statistical analysis; Stock exchanges; Stock realized volatility; Studies; Volatility
• ISSN: 0140-9883; 1873-6181
• DOI: 10.1016/j.eneco.2017.09.023

Academic research relies extensively on stock market information to forecast oil volatility, with relatively little attention paid to the reverse evidence. Our paper fills this gap by investigating the predictive ability of oil volatility risk to forecast stock market volatility. Using oil volatility risk premium (oil VRP) as the predictor, we find that oil VRP does exhibit statistically and economically significant in-sample and out-of-sample forecasting power for G7 countries, even controlling for some popular macroeconomic variables. These findings are robust when using alternative proxies for volatilities of stock and oil. Furthermore, the strength of the predictive evidence is substantial during relatively high and low level of stock market, while is substantially higher for recessions vis-á-vis expansions. Oil VRP can also contains additional information for predicting a series of macroeconomic variables, which serves as an available explanation for its forecasting ability. •Oil volatility exhibits significant forecasting power for G7 stock market volatilities.•Predictability holds well even controlling for some popular macroeconomic variables.•The predictability is robust during relatively high and low level of stock market.•The predictability is substantially higher for expansions vis-á-vis recessions.•Oil volatility contains information for predicting a series of macroeconomic variables.