Backtest MACD Indicator

A Python script to backtest the Moving Average Convergence Divergence (MACD) indicator against the buy-and-hold S&P 500. The backtesting strategy involves long/short position flipping based on the crossover signals. I made this project to learn more about algorithmic backtesting with real market data. Claude.ai was used in the making of this repository

Features

📊 Strategy Implementation

• MACD Indicator: 12/26/9 EMA configuration (Fast EMA, Slow EMA, Signal Line)
• Long/Short Strategy: Automatically flips between long and short positions on MACD crossovers
• Signal Generation: Buy when MACD crosses above signal line, sell short when MACD crosses below
• Trade Log: All trades taken are logged in the console

📈 Performance Analysis

• Alpha Calculation: Measures returns over S&P 500 buy-and-hold
• Risk Metrics: Volatility, Sharpe ratio, and total return calculations
• Benchmark Comparison: Side-by-side performance visualization against market index
• Trade Logging: Detailed console output of all buy/sell/short/cover transactions

🕐 Multiple Timeframes

• Intraday: 5-minute, 30-minute, 1-hour intervals
• Daily/Weekly/Monthly: Standard time periods (Day, Week, Month)
• Custom Date Ranges: User-defined start and end dates

📊 Comprehensive Visualization

Four-panel visual chart:

1. Price Chart: Stock price with buy/sell signal markers
2. MACD Indicator: MACD line, signal line, and histogram
3. Performance Comparison: Strategy vs S&P 500 cumulative returns
4. Metrics Summary: Key performance statistics and alpha interpretation

Installation

Prerequisites

pip install yfinance pandas numpy matplotlib

Required Libraries

• yfinance: Financial data download from Yahoo Finance
• pandas: Data manipulation and analysis
• numpy: Numerical computations
• matplotlib: Plotting and visualization

Usage

Basic Usage

1. Run the script:

python main.py

2. Follow the prompts:
   • Ticker Symbol: Enter any valid stock symbol (e.g., AAPL, TSLA, NVDA)
   • Timeframe: Choose from predefined periods or custom date range
   • Interval: Select candle duration from 5 minutes to 1 month

Example Session

=== MACD Backtest Configuration ===
Enter ticker symbol (e.g., AAPL, TSLA, MSFT): NVDA
Select timeframe (1-7): 4
Select duration (1-6): 4

Downloading data for NVDA...
Downloaded 252 data points
Calculating MACD indicator...
Generating trading signals...
Running backtest...

Strategy Logic

Signal Generation

• Buy Signal: MACD line crosses above signal line → Go long (or cover short and go long)
• Sell Signal: MACD line crosses below signal line → Go short (or sell long and go short)

Position Management

• Long Position: Profit from price increases
• Short Position: Profit from price decreases
• Position Flipping: Automatically switches between long/short on every signal
• Cash Management: Maintains proper cash flow during position transitions

Performance Metrics

• Total Return: Overall strategy performance vs buy-and-hold
• Alpha: Excess return over S&P 500 benchmark
• Volatility: Annualized standard deviation of returns
• Sharpe Ratio: Risk-adjusted return metric

Output

Console Output

• Real-time trade execution logs
• Performance summary with key metrics
• Alpha interpretation (positive/negative)

Visual Dashboard

• Interactive Charts: Price action with signal overlays
• MACD Visualization: Technical indicator with crossover points
• Comparative Analysis: Strategy vs benchmark performance
• Statistical Summary: Comprehensive metrics table

Technical Details

Data Limitations

• Intraday Data: Yahoo Finance limits 5m/30m intervals to the last 60 days
• Market Hours: Intraday data reflects regular trading hours only

Calculation Methods

• MACD: Standard 12/26/9 EMA configuration
• Short P&L: shares_held × (entry_price - current_price)
• Portfolio Value: Dynamic calculation based on current position type (long or short)
• Annualized Metrics: Adjusted for different time intervals

Risk Considerations

⚠️ Important Disclaimers:

• This tool is for educational and research purposes only
• Past performance does not guarantee future results
• Short selling involves unlimited loss potential
• Always consider transaction costs and slippage in real trading
• Backtest results may not reflect real-world trading conditions

Example Results

Positive Alpha Scenario

Strategy Total Return: 15.67%
S&P 500 Total Return: 8.45%
Alpha (Excess Return): 7.22%
✓ POSITIVE ALPHA - Strategy outperformed!

Negative Alpha Scenario

Strategy Total Return: 3.21%
S&P 500 Total Return: 12.34%
Alpha (Excess Return): -9.13%
✗ NEGATIVE ALPHA - Strategy underperformed

Future Updates

• Transitioning to a fully GUI-based program
• Changing the chart style to bar chart
• Adding additional indicators, making a "super backtesting" tool

License

This project is open source

Part of this repository was created with the help of Claude.ai

---

Developed for quantitative analysis and trading strategy research