Backtests Project

Overview

A flexible framework for backtesting trading strategies against historical market data, with a focus on SPY500 comparisons. The system is built with modularity in mind - each strategy exists as an independent module while leveraging shared utilities for data processing, performance metrics, and visualization.

Project Structure

Backtests/
  strategies/             # Strategy implementations
    ma200.py              # 200-day moving average strategy
    monday_reversal.py    # Monday market reversal strategy
    overnight_swing.py    # Overnight holding strategy
    sellmay.py            # Sell in May and go away strategy
    spy_tlt_rotation.py   # SPY/TLT rotation strategy
    turn_of_month.py      # Turn of the month strategy
    berkshire.py          # Berkshire Hathaway vs SPY comparison
  core/
    backtest_utils.py     # Utility functions for data handling, caching, metrics
    plot_utils.py         # Plotting functions for equity curves and stats
    strategy_base.py      # Base strategy interface class
  figures/                # Output figures directory
  main.py                 # Entry point for running strategies
  requirements.txt        # Project dependencies
  README.md               # This documentation

Key Features

• Modular strategy implementation with a common interface
• Efficient data caching to avoid redundant downloads
• Standardized benchmark comparison across all strategies
• Consistent metrics and visualization

How to Run

1. Install dependencies:

   pip install -r requirements.txt
   

2. Run a specific strategy or all strategies:

   # Run a single strategy
   python main.py --strategy ma200
   
   # Run all registered strategies
   python main.py --all
   
   # Get help on available options
   python main.py --help
   

Adding a New Strategy

1. Create a new Python file in the strategies/ directory

2. Implement a class that inherits from StrategyBase with:

   • run(self, data, initial_investment, **kwargs) - Runs the strategy and returns results
   • summarize(self, results, data=None) - Prints summary statistics and creates plots

3. Register your strategy in main.py:

   STRATEGIES = {
       # Add your strategy to this dictionary:
       "your_strategy_key": (
           "strategies.your_module", "YourStrategyClass",
           "SYMBOL", years_back, initial_investment
       ),
       # ...existing strategies...
   }
   

Core Utilities

• Data Management: Automatic caching of downloaded data
• Benchmark Standardization: Consistent benchmark calculations across strategies
• Performance Metrics: CAGR, max drawdown, trade statistics
• Visualization: Equity curve plots with benchmark comparison

Results Output

Each strategy generates:

• Summary statistics in the console (CAGR, drawdown, trade metrics, etc.)
• Comparison to Buy & Hold benchmark (SPY by default)
• Equity curve plots saved in the figures/ directory

License

MIT License