Different Strokes: Return Predictability Across Stocks and Bonds with Machine Learning and Big Data

Abstract

We investigate the return predictability across stocks and bonds using big data and machine learning. We find that machine learning models substantially improve the out-of-sample performance of stock and bond characteristics in predicting future stock and bond returns. Although both stock and bond characteristics provide strong forecasting power for both stock and bond returns, stock (bond) characteristics do not offer significant incremental predictive power above and beyond bond (stock) characteristics in predicting bond (stock) returns. The results also indicate that stock (bond) characteristics are cash flow (discount rate) predictors and stock (bond) return predictability is driven by mispricing (risk) phenomenon.

Keywords: machine learning, big data, corporate bond returns, cross-sectional return predictability

JEL Classification: G10, G11, C13