An Efficient Group Trading Strategy Portfolio Optimization Algorithm

With the growing use of information technology in investment, the group trading strategy portfolio (GTSP) optimization method has been proposed to provide a flexible mechanism for users to select a trading strategy portfolio. A larger solution search area needs to be explored to improve the effectiveness of the existing optimization algorithm for GTSP. However, it presents a computational bottleneck. This paper introduces a parallel genetic algorithm architecture to speed up the GTSP evolution process and proposes an island-based GSTP optimization algorithm. At last, experiments on the real datasets were conducted to prove the proposed approach is significantly faster than the existing approach.