ZenWave validates engine candidates through sequential Walk-Forward Analysis. Each engine maintains its own report, including its WFA slices, robustness analysis, and forward-path comparison.
Long-term USDJPY structural engine · A1–A4 walk-forward slices, timeline, Monte Carlo, live-path overlay.
Open validation report →
Medium-term adaptive engine · B1–B5 walk-forward slices, B5 selection notes, Monte Carlo, live-path overlay.
Open validation report →
Incubating engine · long-side C1–C4 slices published, short side in research, FTMO Swing 2-Step external evaluation.
Open validation report →
Portfolio layer · structure and out-of-sample validation snapshot across A + B under a 0.50% risk cap.
Open validation report →
Walk-Forward Analysis (WFA) is a sequential validation protocol: optimise on in-sample (IS) data, then test on unseen out-of-sample (OOS) segments. Each slice reports performance on data not used by the optimiser.
For ZenWave, WFA is not optional. It is the only accepted validation standard. Published equity curves are stitched from OOS segments.
Validation includes:
This reduces overfitting risk and false robustness.
| Engine | Status | Window | Params |
|---|---|---|---|
| Engine A | Live | Full-window (2014–2025) | Frozen |
| Engine B | Live | B5 slice | Frozen |
As of 31 Dec 2025, Engine A and B parameters are frozen. No parameter, risk, or logic changes are permitted outside a formal retirement or replacement decision.
The live deployment matches expectations set by the combined out-of-sample equity curves:
Research and optimisation are complete. The current phase is observation — tracking live behavior against validated benchmarks.