Examples¶
Python examples covering key chapters from Advances in Financial Machine Learning (AFML) by Marcos Lopez de Prado, implemented using pymlfinance.
Inspired by the BlackArbsCEO AFML Exercises.
Chapters¶
| Chapter | Topic | Polars Expressions |
|---|---|---|
| Ch 2 | Financial Data Structures | — |
| Ch 3 | Labeling | .ml.daily_volatility(), .ml.trend_scanning_label_series() |
| Ch 4 | Sample Weights | — |
| Ch 5 | Fractional Differentiation | .ml.frac_diff_ffd(), .ml.find_min_d(), .ml.adf_test() |
| Ch 6 | Ensemble Methods | — |
| Ch 7 | Cross-Validation | — |
| Ch 8 | Feature Importance | — |
| Ch 10 | Bet Sizing | .ml.sigmoid_bet_size(), .ml.power_bet_size() |
| Ch 11-12 | Backtesting Dangers | — |
| Ch 13 | Synthetic Data | — |
| Ch 14 | Backtest Statistics | .ml.sharpe_ratio(), .ml.hit_ratio(), .ml.compute_drawdowns() |
| Ch 17 | Structural Breaks | .ml.adf_test(), .ml.sadf() |
| Ch 18 | Entropy Features | .ml.binary_encode(), .ml.shannon_entropy(), .ml.lempel_ziv_complexity() |
| Ch 19+20 | Microstructure | .ml.tick_rule_classify(), _lib.vpin(), _lib.kyle_lambda() |
Design¶
- All synthetic data — no external files or API keys required
- Both APIs — each example shows the NumPy API and Polars expression API (where applicable)
- Self-contained — each notebook runs independently
- Exercises — suggested parameter variations at the end of each example