ncBacktester 0.1.0

A simple backtesting engine for trading strategies

ncBacktester

A simple backtesting engine for trading strategies, built for learning purposes.

Overview

ncBacktester is a lightweight Python package for backtesting trading strategies based on hold signals and OHLCV data. Unlike complex backtesting libraries, ncBacktester focuses on simplicity and educational value.

Features

• Simple Strategy Execution: Execute trades based on hold signal changes (0→1 for buy, 1→0 for sell)
• Performance Metrics: Calculate Sharpe Ratio, Sortino Ratio, Annualized Returns, Alpha, Beta, and Max Drawdown
• Stop Loss Support: Fixed and trailing stop loss functionality
• Static Visualization: Simple static plots of equity curves, trades, and drawdowns
• Easy to Use: Clean API similar to backtesting.py but simpler

Installation

pip install ncBacktester

Quick Start

from ncBacktester import Backtest
import pandas as pd

# Your data should have OHLCV columns + Hold_Signal column
data = pd.DataFrame({
    'Open': [...],
    'High': [...],
    'Low': [...],
    'Close': [...],
    'Volume': [...],
    'Hold_Signal': [0, 0, 1, 1, 0, ...]  # 1 = hold, 0 = don't hold
})

# Create and run backtest
bt = Backtest(
    data=data,
    initial_capital=10000,
    stop_loss_pct=0.05,  # 5% stop loss
    commission=0.001  # 0.1% commission
)

results = bt.run()

# View results
print(results['metrics'])

# Plot results
bt.plot()

Requirements

• Python >= 3.8
• pandas >= 1.3.0
• numpy >= 1.20.0
• matplotlib >= 3.3.0

Project Structure

ncBacktester/
├── ncBacktester/
│   ├── __init__.py
│   ├── backtest.py          # Main Backtest class
│   ├── strategy_executor.py # Strategy execution (Coder A)
│   ├── metrics.py           # Performance metrics (Coder B)
│   ├── stop_loss.py         # Stop loss logic (Coder C)
│   └── plotter.py           # Plotting (Coder C)
├── tests/
│   ├── test_strategy_executor.py  # Tests for Coder A
│   ├── test_metrics.py            # Tests for Coder B
│   ├── test_stop_loss.py          # Tests for Coder C (Part 1)
│   ├── test_plotter.py            # Tests for Coder C (Part 2)
│   └── test_integration.py        # Integration tests
├── setup.py
├── pyproject.toml
└── README.md

How It Works

1. Signal Processing: The engine detects when Hold_Signal changes:

   • 0 → 1: Buy signal (go long)
   • 1 → 0: Sell signal (close position)

2. Trade Execution:

   • On buy: Uses available capital to buy as many shares as possible
   • On sell: Sells entire position
   • Prices executed at Close price of the signal bar

3. Stop Loss:

   • Fixed stop loss: Exits if price drops X% below entry
   • Trailing stop loss: Exits if price drops X% below highest price since entry

4. Metrics Calculation: Calculates various performance metrics from the equity curve and trades.

Testing

# Install development dependencies
pip install -e ".[dev]"

# Run all tests
pytest

# Run tests for specific component
pytest tests/test_strategy_executor.py -v  # Coder A
pytest tests/test_metrics.py -v           # Coder B
pytest tests/test_stop_loss.py -v          # Coder C

Development

This project is organized for collaborative development:

• Coder A: Strategy execution and trade management (strategy_executor.py)
• Coder B: Performance metrics calculation (metrics.py)
• Coder C: Stop loss logic (stop_loss.py) and plotting (plotter.py)

Each component has detailed docstrings explaining what needs to be implemented.

Publishing to PyPI

1. Update version in setup.py and ncBacktester/__init__.py
2. Build the package:

   python setup.py sdist bdist_wheel
   

3. Upload to PyPI:

   twine upload dist/*
   

License

MIT License

Contributing

This is a learning project. Contributions welcome!

Acknowledgments

Inspired by backtesting.py but simplified for educational purposes.