Trading Strategy Generation Using Genetic Algorithms

Traditionally, trading strategies and their parameters are heuristically or subjectively constructed by their adopters. Recently, artificial intelligence techniques such as fuzzy logic, neural networks and genetic algorithms are used to solve various problems in trading. This study develops and implements a GA-based trading system which uses genetic algorithms to generate trading strategies based on technical indicators. Parameters of the technical indicators are determined by genetic algorithms using historical data. Simulation results using 4 technical indicators are presented to demonstrate the feasibility and effectiveness of our system.