Natural Gas Trading System

A robust, quantitative natural gas trading system with data pipelines, strategy modules, backtesting, and live trading capabilities.

Project Overview

This project implements a sophisticated natural gas trading system that combines fundamental data (EIA storage, production, consumption), weather data (NOAA forecasts), and technical indicators to generate alpha signals for natural gas futures trading.

Key Features

• Data Ingestion: Automated pipelines for EIA, NOAA, and market data
• Feature Engineering: Weather-based features, storage analysis, and technical indicators
• Strategy Development: Modular alpha, risk, and portfolio construction components
• Backtesting: Vector-based backtesting engine with transaction costs and slippage
• Analytics: Performance analysis and strategy comparison tools
• Live Trading: Signal generation, trade recommendations, and position management

Getting Started

Prerequisites

• Python 3.8+
• Git

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/natural-gas-trading.git
   cd natural-gas-trading
   

2. Create and activate a virtual environment:

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

3. Install dependencies:

   pip install -r requirements.txt
   

4. Set up configuration:

   cp config/credentials.yaml.example config/credentials.yaml
   # Edit credentials.yaml with your API keys
   

API Setup

The system requires the following API keys to function properly:

1. NOAA API - For weather data

   • Register at https://www.ncdc.noaa.gov/cdo-web/token
   • Add your token to config/credentials.yaml

2. EIA API - For natural gas storage and price data

   • Register at https://www.eia.gov/opendata/register.php
   • Add your key to config/credentials.yaml

Project Structure

natural-gas-trading/
├── config/                  # Configuration files
│   ├── config.yaml          # System configuration
│   └── credentials.yaml     # API keys (git-ignored)
├── data/                    # Data storage
│   ├── raw/                 # Raw data from APIs
│   ├── processed/           # Processed datasets
│   └── signals/             # Generated trading signals
├── src/                     # Source code
│   ├── data_ingestion/      # API connectors and data fetchers
│   ├── data_processing/     # Cleaning and feature engineering
│   ├── strategies/          # Trading strategies
│   ├── backtesting/         # Backtesting framework
│   ├── live_trading/        # Live trading components
│   ├── utils/               # Utilities
│   └── visualization/       # Visualizations and dashboards
├── notebooks/               # Jupyter notebooks for analysis
├── tests/                   # Unit and integration tests
├── logs/                    # System logs
└── run_trading_system.py    # Main entry point

Usage

Running Backtests

To run a backtest of the weather-storage strategy over 2 years:

python run_trading_system.py --mode backtest --strategy weather --years 2

To compare all strategies:

python run_trading_system.py --mode backtest --strategy ensemble --years 2

To run individual backtests for all strategies with plotting:

python run_trading_system.py --mode backtest --strategy all --years 2 --plot

Running Live Trading

To generate a single trading signal:

python run_trading_system.py --mode live --once

To run the continuous signal generation service (60-minute intervals):

python run_trading_system.py --mode live --interval 60

System Architecture

Data Pipeline

1. Data Ingestion

   • NOAA weather data fetcher for temperature and precipitation
   • EIA natural gas storage and pricing data
   • Integration with market data sources

2. Feature Engineering

   • Weather features: Heating/Cooling Degree Days (HDD/CDD)
   • Storage deviation from seasonal norms
   • Price momentum and mean reversion indicators
   • Cross-asset signals

Trading Strategies

1. Weather-Storage Strategy

   • Combines temperature anomalies with storage deviations
   • Seasonal adjustment based on time of year
   • Momentum overlay for trend confirmation

2. Momentum Strategy

   • Captures medium-term price trends
   • Volatility-adjusted position sizing

3. Mean Reversion Strategy

   • Identifies overbought/oversold conditions
   • Statistical significance testing

4. Fundamental Strategy

   • Storage-based signals
   • Supply/demand imbalance indicators
   • Seasonal components

Risk Management

1. Position Sizing

   • Volatility-based sizing
   • Maximum position limits
   • Kelly criterion implementation

2. Stop Loss

   • Trailing stops
   • Volatility-based stops
   • Time-based exits

Performance Metrics

• Return Metrics: Total return, annualized return
• Risk Metrics: Sharpe ratio, Sortino ratio, max drawdown
• Trading Metrics: Win rate, profit factor, average win/loss

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

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

Acknowledgments

• NOAA for weather data access
• EIA for natural gas market data
• CME Group for futures market information