Footnote:
Nach Informationen von SSRN wurde die ursprüngliche Fassung des Dokuments June 20, 2022 erstellt
Description:
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