Advanced RustyBT Tutorial Notebooks¶

This directory contains advanced tutorial notebooks covering complex strategies, optimization techniques, and professional trading workflows.

Available Notebooks¶

11. Pipeline API Deep Dive¶

11_pipeline_deep_dive.ipynb

Comprehensive guide to RustyBT's Pipeline API for factor-based strategies: - Factors - Technical indicators, custom computations - Filters - Dynamic universe selection - Classifiers - Categorical asset grouping - Custom Terms - Building proprietary indicators - Production Strategies - Combining multiple factors

Prerequisites: Basic Python, understanding of technical indicators Runtime: ~5 minutes

12. Advanced Order Management¶

12_advanced_order_management.ipynb

Professional order management techniques: - Bracket Orders - Automatic stop loss + take profit - OCO Orders - One-Cancels-Other pairs for breakouts - Trailing Stops - Dynamic stop losses - Order Tracking - Managing order lifecycle - Partial Fills - Handling incomplete executions - Position Scaling - Pyramiding in/out of positions

Prerequisites: Understanding of order types Runtime: ~7 minutes

13. Portfolio Optimization + Walk-Forward Analysis¶

13_portfolio_optimization_walk_forward.ipynb

Combining portfolio allocation with walk-forward validation: - Sub-Strategy Definition - Multiple uncorrelated strategies - Individual Optimization - Walk-forward per strategy - Portfolio Allocation - Fixed and dynamic allocation - Allocation Optimization - Walk-forward optimize weights - Risk Parity - Volatility-based allocation - Kelly Criterion - Growth-optimal allocation - Performance Analysis - In-sample vs out-of-sample metrics

Prerequisites: Notebooks 05 (Optimization), 06 (Walk-Forward), 08 (Portfolio) Runtime: ~15 minutes

14. Multi-Timeframe Strategies¶

14_multi_timeframe_strategies.ipynb

Trading across multiple timeframes for better context: - Timeframe Hierarchy - Higher for trend, lower for entries - Three-Timeframe Alignment - Weekly + Daily + Hourly - Adaptive Timeframes - Switch based on volatility - Higher TF Stops - Use daily structure for stop placement - Timeframe Combinations - Best practices for different styles

Prerequisites: Understanding of technical analysis across timeframes Runtime: ~8 minutes

Recommended Learning Path¶

Path 1: Factor-Based Strategies¶

Start with basic notebooks (01-04)
Notebook 11: Pipeline Deep Dive
Notebook 13: Portfolio Optimization + Walk-Forward
Combine with live trading (09)

Path 2: Discretionary/Technical Trading¶

Start with basic notebooks (01-04)
Notebook 12: Advanced Order Management
Notebook 14: Multi-Timeframe Strategies
Apply to live trading (09)

Path 3: Systematic Portfolio Management¶

Complete basic series (01-10)
Notebook 11: Pipeline Deep Dive (for sub-strategies)
Notebook 13: Portfolio Optimization + Walk-Forward
Deploy with live portfolio allocator

Advanced Topics Covered¶

Quantitative Research¶

Cross-sectional factor analysis
Custom factor development
Factor combination and weighting
Universe selection and screening

Risk Management¶

Bracket orders for automated stops
Trailing stops for profit protection
Position sizing and scaling
Portfolio-level risk parity

Optimization¶

Walk-forward analysis
Parameter robustness testing
Portfolio weight optimization
Adaptive allocation

Execution¶

Order type selection
Partial fill handling
Order modification
Multi-asset order aggregation

Market Analysis¶

Multi-timeframe trend identification
Timeframe alignment strategies
Volatility regime detection
Adaptive strategy selection

Prerequisites¶

Knowledge Requirements¶

Programming: Python (intermediate), pandas, numpy
Trading: Order types, position management, risk management
Statistics: Basic statistics, correlation, optimization concepts
Finance: Portfolio theory, risk metrics (Sharpe, Sortino, etc.)

System Requirements¶

Memory: 8GB+ recommended for optimization notebooks
CPU: Multi-core recommended for parallel optimization
Data: Access to bundle data (yfinance, CCXT, etc.)

Tips for Success¶

1. Start Simple¶

Don't try to combine everything at once. Master each technique individually before combining.

2. Backtest Thoroughly¶

Always backtest advanced strategies across multiple market conditions: - Bull markets - Bear markets - Sideways/choppy markets - High volatility periods - Low volatility periods

3. Paper Trade First¶

Before live trading with advanced techniques: 1. Backtest with realistic transaction costs 2. Paper trade for at least 1-3 months 3. Verify order handling works as expected 4. Monitor for any edge cases

4. Monitor Performance¶

Track these metrics regularly: - In-sample vs out-of-sample performance - Parameter stability over time - Strategy correlation changes - Execution quality (slippage, fills)

5. Risk Management is Paramount¶

Always use stops - Especially with advanced position sizing
Limit position sizes - Don't over-leverage
Diversify strategies - Don't rely on one approach
Monitor drawdowns - Have circuit breakers in place

Common Pitfalls¶

❌ Over-optimization¶

Problem: Curve-fitting parameters to historical data Solution: Use walk-forward analysis, out-of-sample testing, and simple parameter spaces

❌ Ignoring Transaction Costs¶

Problem: Strategies look great in backtest but fail live Solution: Include realistic slippage and commission models

❌ Over-complicating Strategies¶

Problem: Too many parameters, conditions, and timeframes Solution: Keep strategies simple; complexity ≠ profitability

❌ Insufficient Testing¶

Problem: Moving to live trading too quickly Solution: Test across multiple years, market conditions, and assets

❌ Poor Risk Management¶

Problem: Large drawdowns due to lack of stops or over-leveraging Solution: Always use stops, limit leverage, and size positions appropriately

Getting Help¶

Documentation¶

Community¶

Support¶

Check existing issues before creating new ones
Provide minimal reproducible examples
Include error messages and environment details

Contributing¶

Have an idea for a new advanced notebook? We'd love to see it!

Guidelines: 1. Focus on a single advanced topic 2. Include clear explanations and examples 3. Demonstrate with working code 4. Add to this README with proper description 5. Ensure notebook runs without errors 6. Include expected runtime

Potential Topics: - Custom data feed integration - Machine learning factor generation - Options strategies - Futures term structure - Statistical arbitrage - Market microstructure analysis - Risk parity with constraints - Bayesian portfolio optimization

License¶

These notebooks are part of the RustyBT project and are available under the same license terms.