Machine Learning and Dynamic Risk Allocation in Trend Following

This study examines whether a nonlinear machine-learning overlay can improve the risk efficiency of a transparent trend-following strategy across eight international equity ETFs. A conventional twelve-month trend and volatility rule first determines each fund's baseline allocation. A histogram gradient-boosted classifier then makes one decision: retain an eligible sleeve when estimated underperformance risk is acceptable, and hold cash otherwise. A bounded set of model and allocation configurations is ranked on 2010–2015 after-cost certainty equivalent. The selected combination is evaluated from January 2016 through December 2025 using annual chronological refits, fully matured labels, next-close execution, and a three-basis-point one-way trading cost. The overlay earns 4.714% annually versus 4.463% for the baseline, with net Sharpe of 0.519 versus 0.315 and volatility of 4.89% versus 7.88%. Its maximum drawdown is 14.68% compared with 18.48%. The annual certainty-equivalent difference is 1.006 percentage points, with a 95% stationary-bootstrap interval from -1.783 to 3.877 points. In a descriptive January–August 2026 check, the overlay earns 6.289% versus 8.573%, while maximum drawdown is 1.95% versus 10.72%. The evidence shows a stronger historical return-risk profile, although the wide interval and later-period return shortfall limit the claim of persistence.