Skip to content

Quant Strategies

Backtesting and trading framework for crypto and equity markets. Strategies are built around technical indicators (SMA, EMA, RSI, Bollinger Z-score, Stochastic Oscillator) and optimized via N-dimensional grid search over parameter space.

Target: strategies with Sharpe > 1.5 and strong Calmar ratios.

What's Inside

Component Description
Backtest Pipeline data.py → strat.py → perf.py → param_opt.py → walk_forward.py
FastAPI Backend REST + SSE endpoints for optimization, performance, walk-forward
React Frontend MUI-based SPA with configurable factor cards, live progress bar, interactive charts
PostgreSQL Database REFDATA (dropdowns), BT (backtest results), TRADE (live trading) — managed via Liquibase
CLI Full argparse interface for scripted backtests and grid searches

Key Features

  • Multi-factor strategies — combine indicators via AND, OR, or FILTER conjunction
  • SSE-streamed optimization — real-time progress bar with trial count and best Sharpe
  • Data Column selector — backtest on Price or Volume per factor
  • Walk-forward test — in-sample / out-of-sample overfitting detection
  • Paper trading — connect to Futu OpenD for HK/US equity order execution
  • REFDATA-driven UI — all dropdowns sourced from PostgreSQL, zero hardcoded lists