Skip to content

RustyBT Jupyter Notebooks

This directory contains example Jupyter notebooks demonstrating RustyBT's capabilities for interactive backtesting, analysis, and optimization.

Viewing Notebooks

On this documentation site: All notebooks are rendered and viewable in the Notebooks menu (left sidebar) On GitHub: Click the links below to view notebooks in the repository Locally: Download and open in Jupyter Lab/Notebook

Available Notebooks

All 13 notebooks + advanced tutorials are now available! 🎉

⚠️ Note on Multi-Strategy Portfolios: The 09_multi_strategy_portfolio.ipynb has been replaced with a working Python example (portfolio_allocator_tutorial.py). See 09_multi_strategy_portfolio_note.md for the correct implementation pattern.

10_full_workflow.ipynb - Complete end-to-end workflow

Complete workflow demonstrating the entire RustyBT pipeline from data ingestion to optimization.

  • Data ingestion → Strategy development → Backtesting → Analysis → Optimization
  • All major features demonstrated in one comprehensive example
  • Perfect introduction to the framework

🚀 Quick Start Examples

Fast-track examples to get you backtesting immediately:

crypto_backtest_ccxt.ipynb - Cryptocurrency backtesting with CCXT

  • Multi-exchange data fetching (Binance, Coinbase, Kraken)
  • Data validation and quality checks
  • Simple moving average crossover strategy
  • Performance analysis and visualization

equity_backtest_yfinance.ipynb - Stock backtesting with yfinance

  • Stock and ETF data ingestion
  • Strategy development and testing
  • Performance metrics calculation

📚 Getting Started Tutorials

Step-by-step tutorials for learning RustyBT fundamentals:

01_getting_started.ipynb - Your first backtest

  • Setup and configuration
  • Creating your first strategy
  • Running backtests
  • Visualizing results

02_data_ingestion.ipynb - Working with data sources

  • yfinance (stocks, ETFs)
  • CCXT (cryptocurrencies)
  • CSV import
  • Data quality checks

03_strategy_development.ipynb - Building trading strategies

  • Moving average crossover
  • Mean reversion strategies
  • Momentum strategies

📊 Analysis & Optimization

Advanced techniques for strategy evaluation and improvement:

04_performance_analysis.ipynb - Performance metrics deep dive

  • Interactive visualizations with Plotly
  • Key metrics calculation
  • Risk-adjusted returns

05_optimization.ipynb - Parameter optimization

  • Grid search optimization
  • Bayesian optimization
  • Finding optimal parameters
  • Avoiding overfitting

06_walk_forward.ipynb - Walk-forward validation

  • Robust validation techniques
  • Out-of-sample testing
  • Performance degradation analysis

07_risk_analytics.ipynb - Risk metrics and analysis

  • Value at Risk (VaR) calculations
  • Conditional VaR (CVaR)
  • Beta analysis
  • Drawdown analysis

08_portfolio_construction.ipynb - Multi-asset portfolios

  • Equal-weight portfolios
  • Risk-parity allocation
  • Rebalancing logic

09_live_paper_trading.ipynb - Paper trading setup

  • Real-time testing
  • Paper broker configuration
  • Live monitoring

Features Demonstrated

All notebooks showcase: - ✅ Interactive Visualizations using Plotly (equity curves, drawdowns, distributions) - ✅ DataFrame Exports to both pandas and polars formats - ✅ Progress Bars for long-running operations using tqdm - ✅ Async/Await Support for backtests in notebooks - ✅ Rich Display with _repr_html_() methods for strategy objects

Setup

Install Dependencies

# Install RustyBT with notebook support
pip install rustybt plotly tqdm ipywidgets nest-asyncio

# Or using uv (recommended)
uv pip install rustybt plotly tqdm ipywidgets nest-asyncio

Launch Jupyter

jupyter lab

Initialize Notebook Environment

from rustybt.analytics import setup_notebook
setup_notebook()

Usage Patterns

Basic Backtest

from rustybt import TradingAlgorithm
from rustybt.analytics import plot_equity_curve

class MyStrategy(TradingAlgorithm):
    def initialize(self, context):
        context.asset = self.symbol('AAPL')

    def handle_data(self, context, data):
        self.order(context.asset, 100)

algo = MyStrategy(...)
results = algo.run()

# Visualize
fig = plot_equity_curve(results)
fig.show()

Async Backtest with Progress

from rustybt.analytics import async_backtest, setup_notebook

setup_notebook()

# Run async
results = await async_backtest(algo, show_progress=True)

Export to DataFrame

# Export results to polars
results_pl = algo.to_polars(results)

# Export positions
positions = algo.get_positions_df()

# Export transactions
transactions = algo.get_transactions_df()

Visualizations

from rustybt.analytics import (
    plot_equity_curve,
    plot_drawdown,
    plot_returns_distribution,
    plot_rolling_metrics
)

# Equity curve with drawdown
plot_equity_curve(results, show_drawdown=True).show()

# Returns distribution
plot_returns_distribution(results).show()

# Rolling metrics (Sharpe, volatility)
plot_rolling_metrics(results, window=60).show()

Tips for Notebook Development

  1. Always call setup_notebook() at the beginning to configure async support
  2. Use progress bars for operations that take >5 seconds
  3. Export DataFrames for custom analysis with pandas/polars
  4. Save figures for reports: 
    fig.write_html("equity_curve.html")
fig.write_image("equity_curve.png")
  5. Use dark theme for better visibility: plot_equity_curve(results, theme='dark')

Contributing

To add new example notebooks: 1. Follow the naming convention: NN_topic_name.ipynb 2. Include markdown cells explaining each step 3. Demonstrate at least 3 analytics features 4. Add expected runtime in the first cell 5. Ensure notebook runs without errors using jupyter nbconvert --execute

Support

For issues or questions: - Documentation: https://rustybt.readthedocs.io - GitHub Issues: https://github.com/rustybt/rustybt/issues - Discussions: https://github.com/rustybt/rustybt/discussions