CMStatistics 2016: Start Registration
View Submission - CFE
Title: Firm characteristics and the cross section of stock returns: A portfolio perspective Authors:  Victor DeMiguel - London Business School (United Kingdom)
Francisco J Nogales - Universidad Carlos III de Madrid (Spain)
Raman Uppal - EDHEC Business School (United Kingdom)
Alberto Martin Utrera - Lancaster University Management School (United Kingdom) [presenting]
Abstract: More than 300 characteristics have been proposed to explain the cross-section of stock returns. We study which are significant for portfolio construction and why. We highlight three findings. First, without transaction costs five characteristics are significant because they increase mean return, and also reduce the risk of the portfolio of characteristics. A sixth characteristic (beta) is significant only because it reduces risk, even though its mean return is negligible. Prominent characteristics such as momentum and book to market are not significant because their contribution to the portfolio mean is not sufficiently large to offset their contribution to portfolio risk. Second, with transaction costs the number of significant characteristics increases. This is because the rebalancing trades across characteristics offset each other, reducing marginal transaction costs by 65\% relative to those from trading on the basis of single characteristics. Third, exploiting a large set of characteristics results in superior out-of-sample performance, with Sharpe ratio of returns net of transaction costs 150\% larger than that of the value-weighted portfolio, and 100\% larger than that of portfolios that exploit only the three traditional size, momentum, and book-to-market characteristics.