The predictive power of stock style: size, value-growth orientation and the shape of the future return distribution

Abstract: While there has been extensive research trying to explain the variability in the cross-section of expected stock returns, the predictability of other shape characteristics of the future return distribution is a less studied subject. This thesis investigates the relationship between the size and value-growth orientation of stocks (as measured according to Morningstar) and the shape parameters of their future returns. Two particularly interesting results stand out: (i) future volatility tends to be higher for both deep-value and high-growth stocks compared to more moderately valued stocks; (ii) skewness of future returns becomes more negative with stock size. To our knowledge, these empirical findings are not adequately explained by existing theories and therefore highlight the need for further research in this field. Additionally, the predictive model developed in this thesis is tested in a portfolio application exercise. Across different size- and value-growth-specific stock universes, the model correctly identifies portfolios of stocks whose realized performance is consistent with their forecasted risk profile. The results of this exercise suggest the potential practical importance of our empirical findings for portfolio and risk management.