Industry Return Predictability : A Machine Learning Approach

We use machine learning tools to analyze industry return predictability based on theinformation in lagged industry returns from across the entire economy. Controlling forpost-selection inference and multiple testing, we nd significant in-sample evidence ofindustry return predictability. Lagged returns for the financial sector and commodity-and material-producing industries exhibit widespread predictive ability, consistent withthe gradual diffusion of information across economically linked industries. Out-of-sample industry return forecasts that incorporate the information in lagged industryreturns are economically valuable: controlling for systematic risk using leading multi-factor models from the literature, an industry-rotation portfolio that goes long (short)industries with the highest (lowest) forecasted returns delivers an annualized alpha ofover 8%. The industry-rotation portfolio also generates substantial gains during economic downturns, including the Great Recession