Research On Portfolio Selection Problems Based On Emperor Penguin Optimization Algorithm

Portfolio selection is a significant problem in the financial field,which is the optimal allocation of wealth among alternative assets.With the development of financial market and the enhancement of people's investing consciousness,the study of portfolio problems has gotten a lot more attention.The mean-variance model proposed by Markowitz quantifies the expected return and risk of portfolio,opening a new chapter in modern portfolio research.The model,on the other hand,simply simplifies the conditions in the real-world investing environment,and its application in the real-world financial market is still insufficient.As a result,it is necessary to expand the model and put forward the corresponding solution method according to the actual investment situation.Emperor penguin optimization algorithm is a new swarm intelligence algorithm and has not been applied to portfolio problems.The algorithm has simple principle and strong exploitation ability,but has weak exploration ability because of lacking mutation mechanism.In order to balance the exploration and exploitation ability of the algorithm,the original emperor penguin optimization algorithm is improved to enhance the performance of the algorithm in solving portfolio selection problems.In this paper,we design improved emperor penguin optimization algorithms with better performance for different portfolio models.The research contents are as follows:1.A hybrid Emperor penguin algorithm is proposed for portfolio optimization problems with constraints.The performance of the algorithm for solving the portfolio selection problem was enhanced by mixing and improving the emperor penguin optimization algorithm and artificial bee colony algorithm.Experiments on two sets of standard test functions and constrained mean-variance portfolio model are conducted.Compared with other swarm intelligence algorithms verify the robustness of the proposed hybrid algorithm and the great potential for solving constrained portfolio problems.2.An improved Emperor penguin optimization algorithm is proposed for fuzzy portfolio selection problems.Firstly,the feasibility criterion is introduced to deal with the constraints in the fuzzy portfolio model.Secondly,the mutation mechanism is added to balance the exploitation and exploration ability,guide the population to the optimal individual convergence.Finally,Numerical experiments are carried out on standard test function sets and two different fuzzy portfolio problems.The results are compared with those of other swarm intelligence algorithms,the optimization performance of the improved algorithm and its applicability in solving the fuzzy portfolio problem are verified.