首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于高斯核支持向量机和遗传算法的优化组合研究
引用本文:马 静,李星野,徐 荣.基于高斯核支持向量机和遗传算法的优化组合研究[J].经济数学,2017,34(1):11-17.
作者姓名:马 静  李星野  徐 荣
作者单位:上海理工大学 管理学院 数量经济学专业,上海,200093
摘    要:选用2008~2015共8年数据,首先基于高斯核的支持向量机在沪市A股上构建周期性的投资组合,并通过误差图和评价指标与BP神经网络、广义回归神经网络进行比较,结果表明了支持向量机在股票预测上更具有优势.再将改进遗传算法运用于上证股票市场构建最优投资组合,以上证指数作为基准进行比较,得出混合遗传算法优化组合的模型相比单一模型更为有效.

关 键 词:机器学习  高斯核支持向量机  遗传算法  投资组合

Optimal Portfolio Research with Gaussian Kernel Support Vector Machine and Genetic Algorithm
MA Jing,LI Xing-ye,XU Rong.Optimal Portfolio Research with Gaussian Kernel Support Vector Machine and Genetic Algorithm[J].Mathematics in Economics,2017,34(1):11-17.
Authors:MA Jing  LI Xing-ye  XU Rong
Abstract:Based on the Gaussian kernel support vector machine,a portfolio was formed in Shanghai stock market and was compared with the Back propagation neural network and Generalized regression neural network by the relative error figure and some evaluations.The results indicate that support vector machine has the added advantage at stock prediction.The daily price data from January 2008 to December 2015 were used to illustrate the application of support vector machine and genetic algorithm to construct the optimal portfolio in Shanghai stock market, which made a contrast with the benchmark of Shanghai composite index.The results turn out that genetic algorithm optimizes the investment portfolio, and the hybrid model is more effective than a single model.
Keywords:machine learning  Gaussian kernel support vector machine  genetic algorithm  optimal portfolio
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《经济数学》浏览原始摘要信息
点击此处可从《经济数学》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号