A large quantitative model (LQM) that predicts directional price movements for any financial instrument through abstract mathematical reasoning. The model’s core innovation lies in its ability to generalise across all asset classes and time intervals, departing from historical data-constrained methods.
The model was trained exclusively on synthetic price data, a deliberate design choice that eliminates historical bias and overfitting risks. This synthetic-first approach enables real-time adaptation to regime changes and robust handling of volatile market conditions without the need for retraining, while maintaining consistent performance across varying market states.
Combining multi-horizon forecasts, ranging from tick data to weekly timeframes, provides a unified view of price action, enabling dynamic position management and portfolio optimisation through quantified directional signals.
