Momentum and Autocorrelation Patterns in the US Stock Markets

Abstract

This paper seeks to investigate the role of autocorrelation and cross-serial correlation for momentum in stock returns. Using different momentum portfolios applied to the US stock market from January 1941 to December 2021, we find that negative cross-serial correlation drives momentum profits over longer return horizons, while negative autocorrelations act as a reducing factor. However, when the return horizons are shortened, their roles change as autocorrelations become more positive, while cross-serial correlations become less negative. We conclude that underreaction as an explanation of momentum can co-exist alongside negative autocorrelation since the value of serial-correlation varies with different return horizons.