Investment in securities is in an uncertain environment, any gains obtained are accompanied by certain risks. The essence of portfolio optimization is the optimal allocation of the limited assets in securities with different risk and return characteristics. In this paper, the portfolio decision-making utility function is established based on E-SV risk measure, through the analysis of Markowitz portfolio model, an improved portfolio selection criterion is obtained. Because what we solve is a more complex fractional programming portfolio selection model, the traditional algorithm can not guarantee to get global optimum. In view of this, the genetic algorithms of random simulation are introduced to conduct an in-depth study. Empirical analysis shows that the portfolio model and algorithm proposed in this paper is scientific and reasonable, which may provide investors with an effective theoretical guidance and basis for decision making.
Paper
Full text
Research on the Optimal Portfolio Based on Genetic Algorithms
Semantic Scholar · Computer Science · 2013
Abstract
Investment in securities is in an uncertain environment, any gains obtained are accompanied by certain risks. The essence of portfolio optimization is the optimal allocation of the limited assets in securities with different risk and return characteristics. In this paper, the portfolio decision-making utility function is established based on E-SV risk measure, through the analysis of Markowitz portfolio model, an improved portfolio selection criterion is obtained. Because what we solve is a more complex fractional programming portfolio selection model, the traditional algorithm can not guarantee to get global optimum. In view of this, the genetic algorithms of random simulation are introduced to conduct an in-depth study. Empirical analysis shows that the portfolio model and algorithm proposed in this paper is scientific and reasonable, which may provide investors with an effective theoretical guidance and basis for decision making.