Nvestment Portfolio Problem Modeling And Multiobjective Evolutionary Algorithm Research

Investment portfolio problem is a kind of very complicated problem and a well-known NP hard problem with constraints. It is commonly difficult to be solved by traditional methods. Many scholars try to use optimization algorithms, such as quantum evolution algorithm, genetic algorithm, simulated annealing algorithm. More research on portfolio problem is a help to strengthen the risk consciousness of investors and improve the rationality and validity of the decision-making. It is also helpful to reduce the cost of investment, improve the return on investment, and then avoid the blindness of investment.This paper proposes a multiobjective optimization model for investment portfolio problem and solves it by MOEAs. The algorithm shows the application in investment portfolio problem. Finally, computer simulation verifies the feasibility of the model and the validity of the algorithm.In this paper, research work and innovation are given as follows:The research of multi-project and multi-term portfolio problem. There is only one large project discussed in many existed papers,but this situation mostly doesn’t meet the actual investment. Investors might face funding constraints in, and sometimes it is difficult for them to raise funds in one term. Therefore it is useful to consider extending a single term to multi-term and the remaining funds in investment. This paper proposes a multi-project and multi-term portfolio model based on a new kind of Mean-Semi-covariance theory, which describes the uncertainty of return and risk in investment. Multiobjective evolutionary algorithm with greedy repair strategy is used to deal with the infeasible individuals and makes the investment reasonable. We are able to get a series of investment portfolios and the portfolios constitute pareto effective surface. Investors may have different preference to risk and return and then they can choose the portfolio which they preferThe research of securities portfolio problem in the frictionless market.Investors are rational when they face a lot of risky assets in the market. Classical economists conclude that portfolio diversification can reduce risk, but investors increase the number of assets in investment portfolio whether can reduce risk to any low level or to the risk-free, it is a main work of this study. This paper proposes a securities portfolio model with constraints based on Mean-Semi-covariance theory, through considering in the frictionless market. The same algorithm is used to solve this model and makes the investment reasonable. Investors can choose portfolio which they prefer from a series of investment portfolios. In addition, it is proved that when the number of assets in investment portfolio increases at a large size, the risk and return might not be changed and the cost of management might be increased.It is bad for investors to make decision.