Evolving Trading Strategies with Vectorial Genetic Programming

Summary

The document explores vectorial genetic programming (VGP) as a way to evolve financial trading strategies. It introduces two variants: one supports complex-number operations, while the other uses strong typing. Their performance is compared with another VGP variant and standard genetic programming across three financial instruments, using datasets that span more than seven years.

The authors report that profitable strategies could be evolved, and that standard genetic programming ranked among the weakest approaches in every comparison while strongly typed VGP ranked among the strongest. These findings suggest that representation and type constraints can affect the quality of evolved strategies. The description does not identify the instruments, explain the evaluation setup, or report costs, out-of-sample methodology, or risk measures, so it gives limited grounds for judging the strategies’ robustness or live-trading viability.

Key ideas

• Vectorial genetic programming is applied to the evolution of trading strategies.
• One proposed variant supports complex-number operations, and another enforces strong typing.
• The methods are compared with standard genetic programming across three instruments.
• The datasets span more than seven years, and profitable strategies are reported as achievable.
• Standard GP performs relatively poorly in the reported comparisons, while strongly typed VGP performs relatively well.

Tags

# Evolving Financial Trading Strategies with Vectorial Genetic Programming


# Evolving Financial Trading Strategies with Vectorial Genetic Programming









Establishing profitable trading strategies in financial markets is a challenging task. While traditional methods like technical analysis have long served as foundational tools for traders to recognize and act upon market patterns, the evolving landscape has called for more advanced techniques. We explore the use of Vectorial Genetic Programming (VGP) for this task, introducing two new variants of VGP, one that allows operations with complex numbers and another that implements a strongly-typed version of VGP. We evaluate the different variants on three financial instruments, with datasets spanning more than seven years. Despite the inherent difficulty of this task, it was possible to evolve profitable trading strategies. A comparative analysis of the three VGP variants and standard GP revealed that standard GP is always among the worst whereas strongly-typed VGP is always among the best.

Shown in full with attribution under the source's licence. Licence: abstract CC0

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