An improved multi-objective particle swarm optimization for constrained portfolio selection model

This paper addresses the constrained multi-objective portfolio election model for investors by studying three criteria: return, risk, and liquidity. The return rates of securities are assumed to be random variables, the covariances of the return rates of portfolio are adopted to measure the risk and the turnover rates of portfolio are assumed to be fuzzy numbers to measure the liquidity. Then, an improved multi-objective particle swarm algorithm is designed to get a group of non-dominated solutions of the proposed constrained multi-objective portfolio selection model. Finally, a numerical example is also presented to illustrate this algorithm can get a better distribution of solutions.