Changing Expected Returns Can Induce Spurious Serial Correlation

Abstract

Changing expected returns can induce spurious autocorrelation in returns. We show why this happens with simple examples and investigate its prevalence in actual equity data. In a key contribution, we use ex ante expected return estimates from options prices, factor models, and analysts' price targets to investigate our premise. Absolute shifts in expected returns are indeed strongly and positively related to autocorrelations in the cross-section of individual stocks, as predicted by our analysis. Well-studied risk factors show no evidence of spurious components. We also show how our analysis implies spurious cross-autocorrelation and find supporting evidence for this phenomenon as well.