AlphaLens: Institutional Multi-Factor Quant Screener & Regime Pipeline

AlphaLens is an autonomous institutional-grade quantitative research, cross-sectional factor screening, macro regime detection, and portfolio execution terminal. Built on purged walk-forward Machine Learning, statistical mean-reversion modeling, and portfolio-level tail-risk constraints (VaR / CVaR) with automated Alpaca paper broker execution.

---

⚡ Live Quantitative Screener & Terminal

Explore live universe screening, cross-sectional factor rankings, regime classification, and live paper execution telemetry:
👉 Launch AlphaLens Quant Terminal

---

🎯 Key Quant & Engineering Capabilities

• Cross-Sectional Multi-Factor Screener:
  • Live screening across 500+ US equity candidates with real-time sorting by statistical edge, directional probability, 1D return, and RSI.
  • Export full universe factor diagnostics directly to CSV for offline quantitative research.
• Hierarchical Autonomous Intelligence Pipeline:
  • Market Intelligence (market_intelligence.py): Computes log returns, ATR(14), RSI(14), Bollinger Bands, MACD, and real-time sentiment NLP via VADER & TF-IDF on news/social feeds.
  • Quantitative Research (quant_research.py): Fits Ornstein-Uhlenbeck mean-reversion parameters ($\theta, \mu$), simulates Geometric Brownian Motion (GBM) paths, and predicts directional probability via purged walk-forward calibrated Random Forest models.
  • Macroeconomic Regime Scoring (fred_tools.py): Ingests Federal Reserve Economic Data (FRED) including the 10Y-2Y Treasury yield curve spread and VIX to establish systemic risk-on / risk-off factors.
  • Portfolio Risk Council (portfolio_council.py): Solves portfolio-level allocation using half-Kelly criteria, volatility targeting, and strict parametric VaR (95%) and CVaR (95%) gates to veto high-tail-risk candidates.
  • Execution Commander (execution_commander.py): Institutional execution engine enforcing single-point execution, fractional long routing, and TWAP-sliced order execution via Alpaca Markets.
• Automated US Market Session Scheduler:
  • Continuous scheduling via GitHub Actions every 30 minutes during active NYSE/NASDAQ market sessions (EDT / EST adjusted).

---

🏗️ Quantitative Data Flow Architecture

┌────────────────────────────────┐
│   500+ Liquid Candidate Pool   │
└────────────────┬───────────────┘
                 │
                 ▼
┌────────────────────────────────┐       ┌────────────────────────────────┐
│   Market Intelligence Bureau   │ ───>  │  Quantitative Research Bureau  │
│  Technical Factors + NLP Sent  │       │  OU Mean-Reversion + ML P(Up)  │
└────────────────────────────────┘       └───────────────┬────────────────┘
                                                         │
                                                         ▼
┌────────────────────────────────┐       ┌────────────────────────────────┐
│   Execution Commander          │ <───  │   Portfolio Council Risk Gate  │
│   TWAP Slicing & Alpaca Broker │       │   VaR / CVaR & Volatility Cap  │
└────────────────────────────────┘       └────────────────────────────────┘

---

📦 Installation & Setup

1. Clone the repository:

   git clone https://github.com/udbhav-shrinet/Multi-Agent-Trading-Bot.git
   cd Multi-Agent-Trading-Bot

2. Install dependencies:

   pip install -r requirements.txt

3. Configure Environment Variables: Create a config/.env file:

   ALPACA_API_KEY="your_alpaca_key"
   ALPACA_SECRET_KEY="your_alpaca_secret"
   ALPACA_BASE_URL="https://paper-api.alpaca.markets"
   FRED_API_KEY="your_fred_api_key"

4. Run Strategy Pipeline:

   python main.py

---

📄 License

Distributed under the MIT License. See LICENSE for more information.