Nonlinearity in the Cross-Section of Stock Returns: Evidence from China

We study which characteristics provide incremental predictive information for the cross-section of expected returns in the Chinese stock market. Our results provide empirical evidence for strong nonlinear relations between expected returns and selected characteristics, especially in the trading friction category. While a four-factor model of Liu, Stambaugh, and Yuan (2019) explains a majority of anomalous characteristics-sorted portfolio returns, we find significant alphas when exploring these characteristics jointly using flexible predictive functions. A long-short spread portfolio based on out-of-sample predicted returns by a nonlinear model delivers higher Sharpe ratio than that by a linear model. We document more supportive evidence for the nonlinear model after exploring potential interaction effects with firm size, earnings-to-price ratio, and turnover, state dependency of predictors, and various methods of predictive information aggregation, such as forecast combination, principle component regression, and partial least squares.