Stochastic volatility and stocks returns: Evidence from microstructure data.

摘要

Chapter 1 shows that investors' concerns about systematic volatility risk (a distress proxy) can explain some of the small firm premia in the long run. Using a sample of returns spanning the period January 1927-December 2005, we document a statistically significant negative price for volatility risk. This finding indicates a "flight to quality": during recessions, investors shift their preferences away from small firms, which are considered as being relatively distressed. Instead, they use large, "quality" stocks which co-vary positively with innovations in volatility, and therefore pay off during bad economic times, when volatility is more volatile and market returns are lower. This leads to higher prices and lower average returns for large stocks. We find evidence that growth stocks are of hedging value to investors, too. We document a volatility premium ranging from 3% to 7% per year, across value sorted portfolios. We provide assurance that the volatility effect is not subsumed by classical risk factors that is robust with respect to the type of volatility measure used, model specification and that is not sample specific.; Chapter 2 focuses on high-frequency data. The use of high-frequency data in this paper allows us to isolate two potential sources of risk. The first one has to do with the efficient price process. The second one, which is novel, relates to genuine market frictions. We define market friction volatility as the volatility of the difference between observed asset prices and fundamental values.; We propose an asset pricing model having these two intuitively appealing volatility components, market volatility and market friction volatility, and investigate the extent to which it explains documented cross-sectional asset price anomalies.; At monthly frequencies, we find that both volatility measures are negatively priced in the cross-section of stock returns, and they have a strong link to the size anomaly. The small-minus-big premium due to efficient price volatility is as high as 3% per year, while the premium due to market friction volatility is as high as 8% per year, across value-sorted portfolios.; Market volatility and market friction volatility are also negatively priced in the cross-section of daily stock returns. Interestingly, in a 3-factor model comprising both volatility measures and daily market returns, market friction volatility appears to largely subsume the information contained in market volatility. Its premium is as high as 10% per year, across value-sorted portfolios.