Python Trading Platform

Stock analysis and portfolio management platform with ML-powered predictions.

Features

• Stock Analysis - Technical indicators (RSI, MACD, Bollinger Bands) and fundamental metrics (P/E, ROE, debt ratios)
• ML Predictions - Ensemble models using Random Forest, Gradient Boosting, XGBoost, LightGBM, and LSTM neural networks
• Portfolio Tracking - Track holdings, P&L, and performance
• Watchlists - Monitor stocks you're interested in
• Price Alerts - Get notified when stocks hit target prices
• Sector Analysis - Monitor sector ETF performance
• Multi-Exchange - Supports US (NYSE, NASDAQ), Australian (ASX), UK (LSE), and more
• Web Dashboard - Browser-based interface at localhost:8080

Tech Stack

• Python 3.10+
• ML Libraries: scikit-learn, PyTorch, XGBoost, LightGBM
• Data: NumPy, Pandas
• APIs: Alpaca, FMP, Finnhub, Yahoo Finance

Installation

git clone https://github.com/ABurfoot/python-trading-platform.git
cd python-trading-platform

# Required
pip install requests numpy pandas scikit-learn

# Optional - for deep learning
pip install torch xgboost lightgbm

API Keys

Set up API keys as environment variables (all optional - uses available providers):

export ALPACA_API_KEY="your_key"
export ALPACA_SECRET_KEY="your_secret"
export FMP_API_KEY="your_key"
export FINNHUB_API_KEY="your_key"

Usage

Web Dashboard

python3 -m trading.dashboard
# Open http://localhost:8080

Command Line

# Analyze a stock
python3 trade.py analyze AAPL

# International stocks
python3 trade.py analyze ASX:BHP
python3 trade.py analyze LSE:VOD

# Watchlist
python3 trade.py watchlist add AAPL MSFT GOOGL
python3 trade.py watchlist list

# Portfolio
python3 trade.py portfolio summary

Project Structure

python-trading-platform/
├── trading/
│   ├── analyzer.py          # Stock analysis engine
│   ├── dashboard.py         # Web interface
│   ├── data_sources.py      # Multi-API data fetching
│   ├── ml_predictor_v2.py   # ML prediction models
│   ├── indicators.py        # Technical indicators
│   ├── portfolio.py         # Portfolio management
│   ├── watchlist.py         # Watchlist management
│   ├── alerts.py            # Price alerts
│   ├── sectors.py           # Sector analysis
│   ├── backtest_engine.py   # Strategy backtesting
│   ├── risk_manager.py      # Risk calculations
│   └── ...
├── tests/
├── trade.py                 # CLI entry point
└── test_comprehensive.py    # Test suite (277 tests)

ML Models

The prediction system uses an ensemble of:

Model	Library	Type
Random Forest	scikit-learn	Traditional ML
Gradient Boosting	scikit-learn	Traditional ML
Ridge Regression	scikit-learn	Traditional ML
XGBoost	xgboost	Boosting
LightGBM	lightgbm	Boosting
LSTM	PyTorch	Deep Learning
GRU	PyTorch	Deep Learning
CNN-LSTM	PyTorch	Deep Learning

Running Tests

# All tests
python3 test_comprehensive.py

# Quick (skip API calls)
python3 test_comprehensive.py --quick

# Specific category
python3 test_comprehensive.py --category dashboard

Dashboard Tabs

Tab	Description
Analysis	Full technical + fundamental analysis
Compare	Side-by-side stock comparison
ML Predict	Price predictions with confidence intervals
Watchlist	Tracked stocks with live prices
Portfolio	Holdings and P&L
Alerts	Price alert management
News	Stock news feed
Sectors	Sector ETF performance

Disclaimer

For educational purposes only. Not financial advice. Do your own research before investing.

License

MIT