Building an efficient evolutionary algorithm for forex market predictions

Abstract

Foreign Exchange Market is one of the biggest financial markets in the world. In the paper an efficient algorithm for generating profitable strategies on this market is presented. The proposed technique is based on an evolutionary algorithm and uses the combination of technical indicators, which are to enable obtaining the highest profit depending on training and testing data. The algorithm allows to avoid risky strategies and is enhanced in developed mutation and crossover operators for assuring the effectiveness of obtained Forex trade strategies. The performance of the proposed technique was verified by experiments conducted on real data sets.