🤖 ML Models
ML - XGBoost for Financial Prediction
Gradient-boosted decision tree ensemble. The dominant model for tabular (structured) data prediction. In Signal Forge, it predicts BTC direction (up/down in 24h) and altcoin opportunity (expected gain in 24h).
2
Minutes
5
Concepts
+15+30
Read+Quiz
1
How It Works
- Builds trees sequentially — each tree corrects the errors of the previous ones
- Uses gradient descent on a loss function (log-loss for classification)
- Handles missing values, feature interactions, and non-linear relationships natively
- Regularization via tree depth, learning rate, and L1/L2 penalties