关键词: 线性规划, 投资组合模型, 偏度, 交易成本, 有效前沿图像

Abstract: Up to now, there is little research on the model with skewness and transaction cost based on the Markowitz mean-variance model, and there are less research results on this model integrated with Chinese market data and drawing three-dimensional efficiently portfolio frontier images. After setting up two kinds of optimal portfolio models under transaction cost constraints, regarding the linear combination of variance and skewness as objective function, two models are transformed approximately to problems of linear programming by the approximation of linear function. And the approximation degree can be controlled. Then simplex algorithm is applied to solve them to get the optimal portfolios. Depending on these, positive analysis is carried out to utilize the data of the eight Chinese listed companies and to draw approximate three-dimensional efficiently portfolio frontier images. Then the relationships between the objective optimization and parameters are discussed. It can be found that the objective function o is increasing function of expectation r and parameter m.

Key words: linear programming, portfolio model, skewness, transaction cost, efficient frontier images