Time Series Momentum in the US Stock Market: Empirical Evidence and Theoretical Implications
41 Pages Posted: 22 May 2020 Last revised: 24 Feb 2021
Date Written: April 26, 2020
There is much controversy in the academic literature on the presence of short-term trends in financial markets and the trend-following strategy's profitability. This paper restricts its attention to the study of time-series momentum (TSMOM) in the US stock market. The paper aims to suggest answers to several important questions regarding TSMOM and to explain the existing controversy. Our answer to the question, whether short-term trends exist, is strongly affirmative. For the first time, we suppose that the returns follow a p-order autoregressive process with p>1 and evaluate this process's parameters. Fairly accurate knowledge of the momentum generating process allows us to provide analytical results on the profitability of the TSMOM strategy. Our answer to the question, whether the TSMOM strategy is superior to the buy-and-hold strategy, is also affirmative. Finally, we explain the lack of scientific evidence on the TSMOM strategy's profitability. We estimate the power of the statistical test for superiority of the TSMOM strategy and find that the power is much below the acceptable level. Consequently, all empirical tests for the TSMOM strategy's profitability suffer from the low power problem and tend not to reject the false null hypothesis.
Keywords: time series momentum, trend-following, profitability, statistical power
JEL Classification: G12, G14, G17
Suggested Citation: Suggested Citation