Christoph Reschenhofer (Vienna University of Economics and Business)
While the academic literature primarily investigates factor exposures based on covariances (i.e. beta exposure), most practitioners apply characteristics-based scorings to obtain factor portfolios. It hereby remains largely unexplored how firm-level characteristics can be combined to obtain optimal factor portfolios. This paper derives multi-factor portfolios that are formed via a combination of stock characteristic scores. Portfolios that are formed on multiple characteristics are less volatile, and exhibit higher after cost returns compared to the market and single factor portfolios. In addition, return, risk and turnover preferences are very sensitive to buy- and sell-thresholds. We further identify optimal weights for individual factor characteristics, but have to recognize the 1/N factor portfolio as a tough benchmark.
Managing the Market Portfolio
Fabian Hollstein (Saarland U.) and M. Prokopczuk (Leibniz Universität Hannover)
We analyze the relation between time-series predictability and factor investing. We use a large set of financial, macroeconomic, and technical variables to time-series-manage the market portfolio. A combination of the out-of-sample market excess return forecasts of all variables yields a managed market portfolio that generates alphas relative to cross-sectional factor models that exceed 5% per annum. More broadly, the relation between time-series evaluation measures and (multifactor) alphas is weakly positive, but complex. The variables’ predictability for future returns is more important than that for volatility. Finally, we document that managed market portfolios based on lagged factor realizations also perform well.
Multifactor Funds: An Early (Bearish) Assessment
Javier Estrada (IESE Business School)
Multifactor funds, which offer risk factor diversification, have several appealing characteristics. They enable investing in factors, which has become a typical way to enhance a portfolio’s long-term risk-adjusted return; they provide exposure to more than one factor, which enables diversification; and they offer these benefits neatly packaged in one product. What’s not to like? Their performance. Although their track record is limited, the current evidence on multifactor funds targeting the U.S., global, international, and emerging markets shows that these products have largely failed to outperform market-wide, cap-weighted indexes, or low-cost ETFs that track them, in terms of return, risk-adjusted return, and downside protection.
Machine Learning Goes Global: Cross-Sectional Return Predictability in International Stock Markets
Nusret Cakici (Fordham university), et al.
We examine return predictability with machine learning in 46 international stock markets. We calculate 148 stock characteristics and use them to feed a repertoire of different models. The algorithms extract predictability mainly from simple, yet popular, factor types—such as momentum, reversal, value, and size. All individual models generate substantial economic gains; however, combining them proves to be particularly effective. A global value-weighted forecast combination strategy earns 1.51% per month at an annualized Sharpe ratio of 1.49. Despite the overall robustness, the machine learning performance varies substantially across models, countries, and firm size environments. The strategies work best in small stocks, as well as in markets with many listed firms and high idiosyncratic risk limiting arbitrage.
US Structural Drivers of International Portfolio Returns
Bosung Jang (Korea Capital Market Institute), et al.
This paper examines the dynamic response of the global five Fama-French factors and momentum factor to three types of shocks from the US: growth, monetary policy, and risk premium shocks. The results show that although the market portfolio and small-minus-big style respond positively to good news (negatively to adverse shocks), the remaining factors–—high-minus-low, conservative-minus-aggressive, robust-minus-weak, and winning-minus-losing—–are not responsive or respond positively to the negative shocks. Internationally, the factors can respond heterogeneously to the same shock worldwide. By mapping risks to returns, our work paves the way for a detailed exploration of the structural drivers of excess equity returns.
Non-Standard Errors in Asset Pricing: Mind Your Sorts
Amar Soebhag (Erasmus University Rotterdam), et al.
Non-standard errors capture uncertainty due to variation in research design choices. We study the importance of differential design choices in constructing asset pricing factors. By purposely data mining over 250 different versions of each factor, we find that Sharpe ratios exhibit substantial variation within a factor due to different construction choices, which results in sizable non-standard errors and allows for p-hacking. Our study has important implications for model selection exercises.
Learn To Use R For Portfolio Analysis
Quantitative Investment Portfolio Analytics In R:
An Introduction To R For Modeling Portfolio Risk and Return
By James Picerno