quantlab.model.predefined.model_ensemble#
Model ensembles: models of any classes over any factors, combined into one prediction.
ModelEnsemble(members) takes the member models as given, for example an
XGBoost regressor and a GATs network over different factors. Every member
collects its own data and requests its own features; their predictions are
combined label by label: a label several members predict is averaged after a
per-bar cross-sectional z-score (average_predictions), a label one member
predicts is passed through, so a return model and a volatility model make one
predictor. A label several members predict must have one config.
The ensemble satisfies the backtester’s Predictor protocol, so it is
backtested, loaded and rebuilt from a run’s config.json like a single
model.
Classes
An ensemble of models given one by one, of any classes and over any factors. |
|
What a |