A dimension-decreasing particle swarm optimization method for portfolio optimization

Abstract

Portfolio optimization problems are challenging as they contain different kinds of constrains and their complexity becomes very high when the number of assets grows. In this paper, we develop a dimension-decreasing particle swarm optimization (DDPSO) for solving multi-constrained portfolio optimization problems. DDPSO improves the efficiency of PSO for solving portfolio optimization problems with a lot of asset and it can easily handle the cardinality constraint in portfolio optimization. To improve search diversity, the dimension-decreasing method is coupled with the comprehensive learning particle swarm optimization (CLPSO) algorithm. The proposed method is tested on benchmark problems from the OR library. Experimental results show that the proposed algorithm performs well. Copyright is held by the owner/author(s).