Quant Strategy Evolution Loop

A quantitative strategy evolution and backtesting engine for multi-asset classes, designed to simulate professional-grade portfolio management (Citadel-style risk budgeting + Two Sigma volatility targeting).

Core Components

• Backtesting Engine: src/quant_alpha_v1.py, src/quant_backtest_multi.py, etc.
• Auto-Evolution: src/quant_auto_evolution.py (The main "Alpha Evolution Loop" script).
• Risk Management: src/quant_risk_engine.py (Implementation of volatility and drawdown limits).
• Watchdog: scripts/quant_watchdog.sh (Health check and automatic recovery for the background loop).

Strategy Performance Targets (Institutional Grade)

• Sharpe Ratio: > 2.0
• Calmar Ratio: > 2.5
• Max Drawdown: < 8.0%

Asset Classes

• A-Shares (CSI300)
• US Tech (NVDA, etc.)
• Commodities (Gold)
• Crypto (BTC)

Current Status

The system is in continuous "evolution mode," refining factor directions based on historical performance and risk-adjusted returns.