A Limited Attention Theory of Time Series Momentum

Abstract

I present a model of bond and equity markets where some investors have limited attention, and show that the model generates both time series momentum and cross-asset time series momentum. The model makes several novel predictions about the performance of different time series momentum and cross-asset time series momentum strategies, and about the effects that investor attention has on strategy performance. Using bond and equity returns from twenty countries and Google search volume as a measure of investor attention, I find that the predictions are in line with the data.

Keywords: asset pricing, behavioral finance, cross-asset time series momentum, time series momentum, limited attention

JEL Classification: G12, G15, G17, G41, F37