Forecasting global stock market volatilities in an uncertain world

We investigate the predictive relationship between uncertainty and global stock market volatilities from a highfrequency perspective. We show that uncertainty contains information beyond fundamentals (volatility) and strongly affects stock market volatility. Using several crucial uncertainty measures (i.e., uncertainty and implied volatility indices), we prove that the CBOE volatility index (VIX) performs best in point (density) forecasting; the financial stress index (FSI) in directional forecasting. Furthermore, VIX's predictive power improved dramatically after the COVID-19 outbreak, and the VIX-based portfolio strategy enables mean-variance investors to achieve higher returns. There are two empirical properties of VIX: (i) it helps reduce significantly forecast variance rather than bias; and (ii) its forecasts encompass other uncertainty forecasts well. Overall, we highlight the importance of considering uncertainty when exploring the expected stock market volatility.

Automated summary

We examine the relationship between uncertainty and global stock market volatilities using high-frequency data. Our findings show that uncertainty contains valuable information beyond volatility and has a strong impact on stock market volatility. Using various uncertainty measures, we find that the CBOE volatility index (VIX) is the most accurate in point forecasting, while the financial stress index (FSI) performs best in directional forecasting. Additionally, the predictive power of VIX significantly improved after the COVID-19 outbreak, and a VIX-based portfolio strategy allows mean-variance investors to achieve higher returns. Two key empirical properties of VIX are its ability to effectively reduce forecast variance and its superior performance compared to other uncertainty forecasts. Overall, our study emphasizes the importance of considering uncertainty when analyzing expected stock market volatility.