🏦 Autonomous Portfolio Risk Management System

Enterprise-grade autonomous portfolio risk management system with advanced machine learning, real-time monitoring, and comprehensive stress testing capabilities.

https://finaagentai.streamlit.app/

🌟 Key Features

🤖 Multi-Agent Architecture

• Risk Scanner Agent: Real-time volatility and market risk detection
• Analysis Specialist Agent: Deep portfolio concentration and sector analysis
• ML Anomaly Detector: Machine learning-based market anomaly detection
• Stress Test Agent: Comprehensive scenario analysis and stress testing
• Action Executor Agent: Automated risk mitigation actions
• Reporting Agent: Intelligent alert aggregation and reporting

📊 Advanced Analytics

• Risk Metrics: VaR, Expected Shortfall, Maximum Drawdown, Sharpe/Sortino Ratios
• Machine Learning: Isolation Forest anomaly detection with feature engineering
• Stress Testing: Multiple scenario analysis (2008 crash, flash crash, rate shocks)
• Correlation Analysis: Dynamic correlation matrices and risk clustering
• Technical Indicators: RSI, MACD, Bollinger Bands integration

🎛️ Real-Time Dashboard

• Interactive Streamlit Interface: Professional multi-tab dashboard
• Live Market Data: Real-time price feeds and portfolio monitoring
• Risk Heatmaps: Visual risk assessment across assets and sectors
• Alert Management: Comprehensive alert filtering and analysis
• Performance Tracking: Portfolio performance visualization and metrics

🚀 Quick Start

Prerequisites

Python 3.8+
pip (Python package manager)

Installation

1. Clone the repository

git clone https://github.com/YaxitaAmin/FinAgent.git
cd FinAgent

2. Install dependencies

pip install -r requirements.txt

3. Run the dashboard

streamlit run main.py dashboard

4. Access the application

Open your browser to: http://localhost:8501

📦 Dependencies

streamlit>=1.28.0
pandas>=1.5.0
numpy>=1.21.0
yfinance>=0.2.0
plotly>=5.15.0
scikit-learn>=1.3.0
scipy>=1.9.0
requests>=2.28.0
ta>=0.10.2
networkx>=2.8.0
asyncio
sqlite3
threading

🎯 Usage Examples

1. Dashboard Mode (Recommended)

# Launch interactive dashboard
python main.py dashboard

# Or use Streamlit directly
streamlit run main.py dashboard

2. Integrated Mode (Dashboard + Background System)

# Run with background risk monitoring
python main.py integrated

3. CLI Mode (Headless Operation)

# Run command-line version
python main.py

4. Programmatic Usage

from portfolio_risk_system import EnhancedRiskManagementSystem, Portfolio
import asyncio

# Create portfolio
portfolio = Portfolio(
    symbols=['AAPL', 'GOOGL', 'MSFT', 'TSLA'],
    weights={'AAPL': 0.25, 'GOOGL': 0.25, 'MSFT': 0.25, 'TSLA': 0.25},
    total_value=1000000.0
)

# Initialize system
risk_system = EnhancedRiskManagementSystem(portfolio)

# Run analysis
async def main():
    await risk_system.start_system()

asyncio.run(main())

🏗️ System Architecture

graph TB
    A[Real-Time Data Provider] --> B[Risk Management System]
    B --> C[Risk Scanner Agent]
    B --> D[Analysis Specialist Agent]
    B --> E[ML Anomaly Detector]
    B --> F[Stress Test Agent]
    B --> G[Action Executor Agent]
    B --> H[Reporting Agent]
    
    C --> I[Alert Database]
    D --> I
    E --> I
    F --> I
    G --> I
    H --> I
    
    I --> J[Streamlit Dashboard]
    B --> K[Background System Runner]
    K --> J

Loading

📊 Dashboard Overview

Main Tabs:

1. 📊 Overview

   • Portfolio metrics and KPIs
   • Real-time performance charts
   • Risk heatmap visualization
   • System status monitoring

2. 🚨 Risk Alerts

   • Real-time alert feed
   • Alert filtering and search
   • Severity-based categorization
   • Historical alert analysis

3. 📈 Risk Metrics

   • Advanced risk calculations
   • Correlation analysis
   • Volatility tracking
   • Performance attribution

4. 🧪 Stress Tests

   • Scenario-based stress testing
   • Custom scenario creation
   • Impact visualization
   • Recovery analysis

5. 🤖 ML Analytics

   • Anomaly detection results
   • Feature importance analysis
   • Model performance metrics
   • Predictive insights

🔧 Configuration

Portfolio Configuration

portfolio = Portfolio(
    symbols=['AAPL', 'GOOGL', 'MSFT', 'TSLA', 'NVDA'],
    weights={
        'AAPL': 0.20,
        'GOOGL': 0.20,
        'MSFT': 0.20,
        'TSLA': 0.20,
        'NVDA': 0.20
    },
    total_value=2000000.0,
    sector_allocation={
        'Technology': 0.80,
        'Other': 0.20
    }
)

Risk Thresholds

thresholds = {
    'volatility': 0.30,      # 30% annualized volatility
    'var_95': 0.05,          # 5% Value at Risk
    'correlation': 0.80,     # 80% correlation threshold
    'drawdown': 0.15,        # 15% maximum drawdown
    'concentration': 0.25    # 25% single position limit
}

ML Parameters

ml_config = {
    'lookback_period': 30,
    'contamination': 0.1,    # 10% expected anomaly rate
    'confidence_threshold': 0.8
}

🧪 Advanced Features

Custom Stress Scenarios

custom_scenario = {
    'equity_shock': -0.35,     # 35% market decline
    'vol_spike': 3.0,          # 3x volatility increase
    'correlation_increase': 0.3, # Correlation spike
    'recovery_time': 30        # Days to recovery
}

Machine Learning Features

• Technical Indicators: RSI, MACD, Bollinger Bands
• Volume Analysis: Volume-price relationships
• Momentum Indicators: Price momentum and trend analysis
• Volatility Metrics: Rolling volatility and GARCH modeling

Risk Metrics Calculations

# Value at Risk (VaR)
var_95 = AdvancedRiskMetrics.calculate_expected_shortfall(returns, 0.05)

# Maximum Drawdown
max_dd = AdvancedRiskMetrics.calculate_max_drawdown(prices)

# Sharpe Ratio
sharpe = AdvancedRiskMetrics.calculate_sharpe_ratio(returns)

# Sortino Ratio
sortino = AdvancedRiskMetrics.calculate_sortino_ratio(returns)

📈 Performance Optimization

Data Caching

• Intelligent market data caching
• Database query optimization
• Memory-efficient data structures

Asynchronous Processing

• Non-blocking I/O operations
• Concurrent agent execution
• Background system monitoring

Resource Management

• Automatic memory cleanup
• Connection pooling
• Efficient data serialization

🛡️ Risk Management Features

Alert Severity Levels

• CRITICAL: Immediate action required
• HIGH: Urgent attention needed
• MEDIUM: Monitor closely
• LOW: Informational
• INFO: System notifications

Automated Actions

• Position size adjustments
• Hedge recommendations
• Rebalancing suggestions
• Stop-loss triggers

Compliance & Monitoring

• Real-time position monitoring
• Regulatory limit tracking
• Audit trail maintenance
• Performance reporting

🔍 Monitoring & Alerts

Real-Time Monitoring

# Portfolio monitoring
monitor_portfolio_value()
track_risk_metrics()
detect_limit_breaches()

# Market monitoring
scan_volatility_spikes()
detect_correlation_changes()
monitor_sector_rotation()

Alert Types

• HIGH_VOLATILITY: Excessive asset volatility
• CONCENTRATION_RISK: Position size violations
• SECTOR_CONCENTRATION: Sector allocation limits
• MARKET_ANOMALY: ML-detected anomalies
• STRESS_TEST_FAILURE: Failed stress scenarios
• CORRELATION_SPIKE: High correlation events

📊 Data Sources

Market Data

• Yahoo Finance: Real-time price data
• Technical Analysis Library: Indicator calculations
• Custom APIs: Alternative data sources

Risk Data

• Historical Prices: Volatility calculations
• Volume Data: Liquidity analysis
• Corporate Actions: Dividend adjustments
• Economic Indicators: Macro risk factors

🚀 Deployment Options

Local Development

streamlit run main.py dashboard

Docker Deployment

FROM python:3.9-slim

WORKDIR /app
COPY . .
RUN pip install -r requirements.txt

EXPOSE 8501
CMD ["streamlit", "run", "main.py", "dashboard"]

Cloud Deployment

• Streamlit Cloud: Easy one-click deployment
• AWS ECS: Container-based deployment
• Google Cloud Run: Serverless deployment
• Azure Container Instances: Cloud container hosting

🤝 Contributing

We welcome contributions! Please see our Contributing Guidelines for details.

Development Setup

# Clone the repository
git clone https://github.com/YaxitaAmin/FinAgent.git

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install development dependencies
pip install -r requirements-dev.txt

# Run tests
pytest tests/

# Run linting
black . && flake8 .

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Yahoo Finance API for market data
• Streamlit for the amazing dashboard framework
• Plotly for interactive visualizations
• scikit-learn for machine learning capabilities
• Technical Analysis Library for financial indicators

📞 Support & Contact

• Email: yaxita2003@gmail.com

---

Built with ❤️ for the financial technology community

---

⚠️ Disclaimer: This system is for educational and research purposes. Always consult with qualified financial professionals before making investment decisions.