Every downstream decision needs to know how much to trust the forecast. HELIOS reports that trust as a calibrated range, not a single figure.
It doesn't just average its models; it accounts for when they're wrong at the same time. Covariance-aware ensemble weighting.
The forecast range is calibrated, so a stated confidence level is the one you actually get. Locally-adaptive calibrated (conformal) prediction intervals.
The band narrows in calm conditions and widens when the market is hard to call. Locally-adaptive interval width.
STRIX simulates over this range; PHALANX uses it to see the peak coming. Provides the forecast distribution to STRIX and PHALANX.
Our validation methodology is a walk-forward backtest against standard baselines with Diebold–Mariano significance testing — designed to run on your own history during onboarding so you can see how it performs before you rely on it.
These are the real methods behind HELIOS, each in one plain line. We name the method as a credibility signal — we don't publish the recipe.
Blends its models by how they err together, not a flat average.
Gives a range calibrated to real outcomes, tighter when confident.