How Trend Signals Relate: Time-Series Momentum and Moving Average Crossovers
Summary
This document explains how time-series momentum (TSMOM) and moving-average crossover (MACROSS) strategies can represent closely related ways to measure price trends. It describes expressing trend signals through past prices or returns, and using “signature” plots to show how a signal weights historical information. Exponential moving averages, HP and Kalman filters, and OLS trend slopes are also presented as filters that can be analyzed within this framework.
For an empirical comparison, the authors pair TSMOM lookbacks spanning one, three, and twelve months with MACROSS settings chosen to produce comparable trend measures. The document reports that corresponding strategy pairs have similar returns and that the tested signals show excess returns. It provides no detailed performance statistics in the text, and the findings do not establish that trend filters will work equally well across markets or periods. Its practical conclusion is to give substantial attention to risk control, portfolio construction, position sizing, and execution.
Key ideas
- Time-series momentum and moving-average crossovers can encode similar trend information.
- Trend signals can be decomposed into their weights on historical prices or returns.
- Signature plots offer a way to compare the information used by different trend filters.
- The reported empirical comparison finds similar returns for paired momentum and moving-average signals.
- Risk controls, portfolio construction, position sizing, and execution also matter to CTA strategy results.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.