A Holistic Approach to the Predictive Power of Expected Volatility

Abstract

This paper studies the predictive power of expected volatility in the cross-section of expected stock returns. Evidence indicates that total and idiosyncratic volatility levels and volatility innovations have predictive power in the cross-section of expected excess stock returns. The results show that volatility levels are positively and volatility innovations are negatively related to future stock returns. EWMA, GARCH and TGARCH volatility measures have the largest predictive power, whereas the predictive power of realized and implied volatility measures is mixed. Additionally, we control the results for the short-term reversal effect and illiquidity, but neither of those effects helps explaining the predictive power of expected volatility.

Keywords: Cross-sectional returns, idiosyncratic volatility, GARCH, implied volatility

JEL Classification: C53, G11, G12, G17