基于SVM-GRA-模糊熵的项目组合风险优化  

Optimization of portfolio risk based on SVM-GRA-fuzzy entropy

作　　者：杨菊欢 郑唯唯[1] 段博文[1] Yang Juhuan Zheng Weiwei Duan Bowen(College of Science, Xi＇an Polytechnic University, Xi＇an 710048, Shaanxi Province, P. R. China)

机构地区：[1]西安工程大学理学院,陕西西安710048

出　　处：《深圳大学学报（理工版）》2016年第6期586-592,共7页Journal of Shenzhen University(Science and Engineering)

基　　金：国家自然科学基金资助项目(71272049)~~

摘　　要：针对交互效应下项目组合风险的不确定性和不对称性特点,构建项目组合风险优化模型.借助支持向量机筛选符合战略目标的项目,利用资源、技术及收益的三角模糊数及其熵值反应不确定性,采用灰色关联度刻画三维不对称交互效应,获得项目组合风险测度.基于模糊规划理论,得到满足期望收益下的项目组合风险优化策略,利用数值模拟验证了方法的有效性.A novel project portfolio risk optimization model is proposed based on the uncertainty and asymmetry of the portfolio risk under interactive effects. Specifically,the projects in line with the strategic target are filtered using the support vector machine（ SVM）. The uncertainty is reflected by the triangular fuzzy number and entropy of portfolio＇s resources, technology and benefits. In order to measure the portfolio risk, the three-dimensional asymmetric interactive effect is depicted by employing the grey relational degree. According to the fuzzy planning theory,the optimization strategy which satisfies the expected return of portfolio risk is obtained. Last, the effectiveness of the method is validated by utilizing numerical simulation,and the approach provides a powerful basis for policymakers to measure the portfolio risk and choose the appropriate project portfolio.

关 键 词：计算数学 项目组合风险 交互效应 支持向量机 灰色关联分析 模糊熵 

分 类 号：O29[理学—应用数学] F272.3[理学—数学]