Abstract. Due to the important effect of the transaction cost, risk, skewness and kurtosis to portfolio returns, the aim of this paper is to simulate the real transactions in stock market by considering the above factors. Firstly, two mean-semi-variance-skewness-kurtosis portfolio optimization models in open-loop and closed-loop are proposed by considering the transaction cost, return, risk, skewness and kurtosis. Secondly, the fuzzy programming approach is used to transform the two models into the corresponding single-objective programming models, and the genetic algorithm with adaptive scale adjustment is designed to solve them. Finally, the real data from the Shanghai Stock Exchange is given to illustrate the advantage of the proposed models and the efficiency of the designed algorithm.