A Divide-and-Conquer-based Approach for Diverse Group Stock Portfolio Optimization Using Island-based Genetic Algorithms
Chun-Hao Chen, Wan-Yi Shen, Mu‐En Wu, Tzung‐Pei Hong · 2019
Because portfolio management is a problem of optimization, many approaches have been presented to address it. In this paper, we propose an enhanced approach to obtain a diverse group stock portfolio (DGSP) from a large number of stocks using the island-based grouping genetic algorithm with the divide-and-conquer strategy. The main concept is that the proposed approach divides the given stocks into several subsets, and for every island, it will select a random subset to form its own initial population. During the evolution process, the stock synchronization mechanism is designed to adjust stocks in an island after chromosome migration. At last, experimental results on a real dataset were made to show the effectiveness and efficiency of the proposed approach.