Multi-period and tri-objective uncertain portfolio selection model: A behavioral approach

Abstract

This paper discusses a multi-period and tri-objective portfolio optimization problem, where asset returns are formalized as uncertain variables. Based on uncertainty theory, a multi-period and tri-objective uncertain portfolio model is proposed, which considers the loss-averse utility, liquidity risk and diversification degree simultaneously. Additionally, a chance constraint is introduced into the model to reflect investors’ safety requirement during the investment period. To solve the portfolio model, a self-adaptive particle swarm optimization (SAPSO) is also proposed. In SAPSO, a self-adaptive stochastic ranking approach is employed to balance the abilities of exploration and exploitation in the searching process. Finally, a numerical experiment is presented. The results show that SAPSO is effective to solve the proposed model and the proposed portfolio model can express investors’ preference by adjusting the objective weights.