Heuristic Algebra and LLM Assisted Genetic Algorithms for Trading Strategy Development
A research project applying heuristic algebra and LLM-assisted genetic algorithms to trading strategy development. The paper frames strategy evolution as search over heuristic-algebra genomes, then studies the approach empirically across markets.
Full paper: PDF
What it is
The paper is a 16-page cross-market empirical study combining:
- heuristic algebra
- LLM assistance
- genetic algorithms
- quantitative trading
- regime detection
- walk-forward evaluation
- futures-market strategy development
The core idea is to evolve trading strategies over structured heuristic-algebra representations rather than relying only on conventional parameter search.
Why it matters
This project sits directly at the junction of Phase 3 and Phase 4:
- Phase 3 — trading strategy design, futures markets, systematic research
- Phase 4 — heuristic algebra as a formal system for structured search and reasoning
It extends heuristic algebra from prompt scaffolds into trading strategy development and empirical market testing.
See Also
- phase-3-trading-algorithms-and-hedge-funds — trading and research background
- heuristic-algebra — the formal system used in the paper
- heuristic-algebra-paper — earlier heuristic algebra paper
- llm-ga-heuristic-algebra-paper — source page for this PDF