Stock Return Serial Dependence and Out-of-Sample Portfolio Performance

Abstract:     
We study whether investors can exploit stock return serial dependence to improve the out-of-sample performance of their portfolios. To do this, we first show that a vector autoregressive (VAR) model estimated with ridge regression captures daily stock return serial dependence in a stable and statistically significant manner. Second, we characterize (analytically and empirically) the expected return of an arbitrage (zero-cost) portfolio based on the VAR model, and show that it compares favorably to that of other arbitrage portfolios in the literature. Third, we evaluate the performance of two investment (positive-cost) portfolios: a conditional mean-variance portfolio obtained using the linear VAR model, and a conditional mean-variance portfolio using a nonparametric autoregressive (NAR) model. We show that, subject to a suitable norm constraint, these two investment portfolios outperform the traditional (unconditional) portfolios for transaction costs below 10 basis points.

Number of Pages in PDF File: 58

Keywords: Serial dependence, vector autoregression, portfolio choice, out-of-sample performance

JEL Classification: G11

working papers series