A new multi-objective portfolio optimization model based on dual expected utility

This paper gives a new portfolio optimization model based on dual expected utility. In the model, considering that the transaction number is integer and short sale is not allowed in China´s Stock Market at present, we introduce the constraints of minimal transaction unit and upper bound of investing in each asset in the friction market and we regard the utility risk as the risk under the non-conventional expected utility theory. Under the condition that the yield of portfolio has normal distribution, we propose a new portfolio optimization model. The new model is a nonlinear integer programming problem, we propose an improved genetic algorithm for solving the problem and it is shown with numerical result that the given model is reasonable and the given algorithm is efficient.