signal-classification

What problem does it solve?

This Skill automates the prediction of whether an asset's price will move up or down over a future period using machine learning, enabling more informed trading decisions.

Core Features & Use Cases

• Predictive Modeling: Utilizes XGBoost and LightGBM for accurate price movement forecasts.
• Robust Validation: Employs walk-forward validation to prevent lookahead bias and ensure realistic performance.
• Interpretability: Provides SHAP feature importance to understand the drivers behind predictions.
• Threshold Optimization: Fine-tunes prediction thresholds to maximize trading profitability.
• Use Case: A quantitative trader can use this Skill to build a model that predicts the probability of a 1% price increase in BTC/USD within the next 24 hours, helping to identify potential long entry signals.

Quick Start

Train a signal classification model using the provided features and labels, applying walk-forward validation.