Dice Question Streamline Icon: https://streamlinehq.com

Independence versus redundancy of time-scale layers in multi-horizon trend-following

Determine whether each lookback horizon used in multi-horizon trend-following strategies (e.g., 20-, 60-, 125-, 250-, and 500-day windows) provides independent diversification when combined, or whether some horizons are redundant because they capture information already embedded in adjacent horizons; assess this specifically within the context of managed-futures (CTA) implementations that blend short-, medium-, and long-term signals.

Information Square Streamline Icon: https://streamlinehq.com

Background

Trend-following CTAs commonly combine multiple lookback windows to diversify across time scales. While this practice is widely believed to enhance robustness, the authors note that empirical work has rarely isolated whether each horizon contributes distinct diversification once adjacent horizons are included.

The paper questions whether intermediate horizons, especially in the medium-term range, add independent value or simply overlap with short- and long-term components. Establishing independence versus redundancy is central to understanding true time-scale diversification in CTA systems.

References

Yet it remains unclear whether each layer provides independent diversification or whether some horizons are redundant—capturing information already embedded in adjacent ones. This open question lies at the intersection of two strands of literature: the study of trend premia as a systematic return source, and the broader portfolio-theoretic discussion of how diversification interacts with redundancy in correlated strategies.

— Revisiting the Structure of Trend Premia: When Diversification Hides Redundancy (2510.23150 - Etiennea et al., 27 Oct 2025) in Section “Background and Literature Review,” Subsection “Diversification across time scales”