I’ve been exploring a different approach to market analysis by embedding trade volume into an ordinary differential equation (ODE) computed directly in SQL with QuestDB. Instead of relying on correlation, the method treats volume as a dynamic variable projected through cosine and sine transforms inside the database, revealing a geometric phase relationship between price and volume at the tick level.

Highlights: • Implemented fully in QuestDB SQL + Grafana • Live Binance XRPUSDT data • The database itself performs the ODE transform • Shows a strong geometric coupling that vanishes at aggregated scales