QuantGplearn

A systematic framework for factor mining in quantitative investment strategies

Improvements Made in This Project

This project introduces the following improvements:

1. Added new temporal operators as listed below, with extensive vectorization and Numba optimization for better performance. A complete list of these operators can be found here: https://zhuanlan.zhihu.com/p/24627909174.
2. The original gplearn package allowed setting constant parameters only within a single range. In this project, we enhanced this functionality by enabling the selection of constants from a set. This change is particularly useful for setting parameters based on economic logic, such as selecting quantization strategies for 1-day, 7-day intervals, etc.
3. Added a penalty term to the fitness function that limits the length of formulas, preventing formulas from growing indefinitely during iterations.
4. Introduced a maximum length restriction: if the length exceeds a certain threshold, no crossover, subtree mutation, point mutation, or point replacement will occur. Only pruning mutations will be applied.
5. Optimized the slow computation of all factors by utilizing pathos.multiprocessing for parallel processing.

Example Code

Example code for this project can be found in example/get_factors.py. Running this script will generate the following example factors:

[ts_min(ts_atr(zscore_7d, high, mom_3d, 24, 24), 72), ts_macd(ts_hedge(ts_bar_bs(ts_zscore(ts_cdlbodym(zscore_7d, zscore_1d, 24), 72), ts_cmo(abs(mom_1d), 72), 24), ts_kurt(mul(ts_mom(ma_7d, 72), ts_delta(ma_1d, 72)), 72), 72, 24), 24, 24, 72)]

Acknowledgements

I would like to express my gratitude to the developers of the following projects, whose work has significantly contributed to the development of this package:

• gplearn: This repository provided foundational ideas and code that inspired the design and implementation of this package.
• gplearnplus: I used this repository as a reference for extending genetic programming functionalities in my project.

Thank you to both projects for their valuable contributions!