Financial Glossary
Ensemble
Instead of a single model, LearnAImarkets combines several. Like asking multiple experts and going with the majority.
Explanation
Ensemble of Transformers trained with three different fixed seeds, plus XGBoost. The Transformer average is blended with XGBoost per horizon, weighted by the out-of-sample error measured in walk-forward.
Technical detail
The XGBoost weight is not fixed: it starts at 0.40 and moves between 0.15 and 0.85 based on its OOS quality per horizon, with a gate that caps it if its MAE is more than 10 % worse than the Transformer.
See also
Educational content. Not investment advice. Past results do not guarantee future returns.