Quantitative XAU/USD Session Strategy A quantitative research project to model and predict the session dynamics of Gold (XAU/USD). This repository explores multiple hypotheses using a dual modeling approach, benchmarking Gradient Boosted Trees (XGBoost) against Deep Learning models (LSTM, Transformers) for each research question. The project investigates both classification (predicting market direction) and regression (predicting market return) to build a comprehensive, backtested understanding of intra-day market behavior. Project Goals & Research Questions The primary goal of this project is to move beyond simple discretionary trading patterns and answer key quantitative questions through a rigorous, data-driven framework: 1. Predictive Power of Sessions: Can the characteristics of one trading session (e.g., Asia) reliably predict the behavior of a subsequent session (e.g., London or New York)? 2. Intra-day Momentum: Can the price action of the London morning session be used to forecast the high-volume London/New York overlap? 3. Model Benchmarking: Which model architecture—XGBoost

⚠ No verified equity curve — no track-record source connected.

No score history is stored yet — only the current score is shown.

⚠ No live verification account connected — ask for proof before buying.

Alerts on changes: coming soon

Prop-firm compatibility not provided.

Data-completeness & trust index (not a profitability rating)