Dynamic adjustment of horizon exposures versus static equal weighting

Investigate whether dynamically adjusting asset-level weights assigned to trend lookback horizons through time improves robustness compared with static equal weighting in multi-horizon trend-following strategies; evaluate the efficacy of time-varying horizon allocations relative to fixed, equal-weight schemes.

Background

Most academic and practitioner implementations combine multiple horizons using static or equal weighting to avoid overfitting, implicitly assuming that horizon importance is time-invariant across assets.

The authors highlight the unresolved question of whether adaptively reweighting horizons over time can produce more robust performance, motivating their Bayesian optimization framework to test dynamic horizon allocation.

References

Despite a rich literature on trend-following and time-horizon diversification, one critical question remains open: can horizon exposures be dynamically adjusted over time to improve robustness, rather than relying on static equal weighting?

— Revisiting the Structure of Trend Premia: When Diversification Hides Redundancy (2510.23150 - Etiennea et al., 27 Oct 2025) in Section “Background and Literature Review,” Paragraph “Limitations of the literature and contribution of this study”