Optimization of Investment Portfolio Based on Improved Multi-Objective Genetic Algorithm

Abstract

The securities market is characterized by high returns and high risks, and the issue of securities investment portfolio has always been a problem worthy of study. Multi-objective genetic algorithm is widely used in portfolio problems because of its ability to deal with large-scale search space independently of the problem domain, and to solve the problem through loop iterative parallel search. However, the local search performance of the existing multi-objective genetic algorithm is relatively weak, and the cross-mutation process ignores the density information around the individual, which limits the search performance of the algorithm to a certain extent. In order to solve the above problems, this paper proposes an improved multi-objective genetic algorithm investment portfolio scheme. First, the improved Sigmoid function is introduced to realize the adaptive change of the mutation operator, and the population distance between individuals is incorporated into the cross-mutation operation to optimize the search performance of the algorithm. A large number of experiments show that this scheme can be used to solve the Pareto optimal solution set of the portfolio optimization problem, and it has faster convergence than the multi-objective genetic algorithm before the improvement, which can effectively improve the Pareto optimization of the securities investment portfolio. The search performance of the solution set.