Using Genetic Programming to Discover Predictive Trading Factors

Summary

This brief overview introduces genetic programming as an evolutionary approach to generating quantitative investment factors. It describes starting with basic data inputs and mathematical operators, then iteratively evolving candidate expressions through mechanisms inspired by natural selection. The intended goal is to find factors with predictive ability. The page points readers to accompanying educational slides and separate example implementations for stocks and futures.

The document offers a high-level description rather than a worked method: it does not specify the input data, fitness function, selection process, validation design, or portfolio construction. It also reports no predictive results or comparisons. In practice, a factor discovered through repeated search would need careful out-of-sample evaluation and controls for overfitting, but those safeguards are not discussed in the supplied text. The material is therefore useful as an introduction to automated factor discovery, while leaving the implementation and evidence to the linked resources.

Key ideas

• Genetic programming can evolve candidate trading factors from data inputs and mathematical operators.
• The process uses evolutionary mechanisms to search for expressions with predictive potential.
• The page points to separate stock and futures examples and educational slides.
• It does not describe validation methods or provide evidence of predictive performance.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.