中文摘要：

在考虑资产收益率分布中正的偏度水平前提下，以风险价值VaR为约束条件，并引入非线性交易费用、税收等市场摩擦因素，建立以累积偏度最大为目标函数的多期投资组合优化模型，用罚函数法和PSO算法结合求解此模型，并给出实证分析。考虑到在买卖资产风险时交易费用等对投资收益的影响，投资者应该在每一期都对其资产组合进行调整分析，确保在每一期的开始都建立起符合需要的最优资产组合，这对投资者的连续投资行为具有一定的指导意义。

英文摘要：

The paper sets up a multi - period portfolio optimization model that takes maximum of cumulate skewness as objective function by using risk measure index VaR and introducing non - linear transaction cost and revenue in the context of the level of positive skewness of return on assets distribution. We combine the dynamic punished function method with PSO algorithm to solve the problem. Finally, a numerical example is given to test the model. Considering the impact of transaction costs on investment income in buying and selling risk assets , in each period, the investor should adjust their asset portfolio to ensure construct the optimal asset portfolio at the beginning, which is helpful for the investor＇ s continuous investment.