Optimization of technical trading strategy using Split Search Genetic Algorithms

Recently, the use of genetic algorithms for the optimization of technical trading strategies has been receiving a great deal of attention. Studies by Alien and Karjalainen (1994) and Bauer (1994) have shown genetic algorithms are capable of developing extremely profitable technical trading strategies. This paper looks at the uses of genetic algorithms in the optimization of parameter values in a technical trading strategy and proposes a novel hybrid genetic algorithm, focused on the role of mutation in the evolutionary process. The novel Split Search Genetic Algorithms (SSGAs) is assessed in performance against STanDard Genetic Algorithms (STDGAs). Results from preliminary function optimization show that this novel algorithm is more effective and efficient in locating the optimal solution. The profitability of technical trading strategies optimized with SSGAs is greater than those optimized with standard GAs.