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

GA在改善证券市场风险测量精度中的实证研究
引用本文:盖卉,张磊,别晓竹.GA在改善证券市场风险测量精度中的实证研究[J].哈尔滨商业大学学报(自然科学版),2007,23(3):372-377,381.
作者姓名:盖卉  张磊  别晓竹
作者单位:1. 中山大学,数学与计算科学学院,广东,中山,510275
2. 北京理工大学,管理与经济学院,北京,100081
摘    要:应用遗传算法(GA)估计线性GARCH(1,1)模型和非线性EGARCH(1,1)的参数,建立两个模型计算中国证券市场的风险值(VaR).通过对比传统算法和遗传算法的风险值,计算数据显示:线性GARCH(1,1)模型和非线性EGARCH(1,1)模型下,GA得到的风险值均大于真实值,可以解决传统算法中预测精度不高的问题.

关 键 词:遗传算法  BHHH算法  VaR  GARCH  EGARCH
文章编号:1672-0946(2007)03-0372-06
收稿时间:2006-10-23
修稿时间:2006-10-23

Value at risk based on genetic algorithm and BHHH algorithm: an empirical study of Chinese Stock Markets
GAI Hui,ZHANG Lei,BIE Xiao-zhu.Value at risk based on genetic algorithm and BHHH algorithm: an empirical study of Chinese Stock Markets[J].Journal of Harbin University of Commerce :Natural Sciences Edition,2007,23(3):372-377,381.
Authors:GAI Hui  ZHANG Lei  BIE Xiao-zhu
Institution:1. School of Mathematics and Computational Science, SUN YAN-SEN University, Zhongshan 510275, China; 2. School of Management and Economics, Beijing Institute of Technology, Beijing 100081, China
Abstract:Genetic algorithm(GA) has been employed to improve the accuracy of parameters estimation in the linear model GARCH(1,1) and the non-linear model EGARCH(1,1).And models have been constructed to calculate Value at Risk(VaR) from the Chinese stock markets.As compared with VaR gotten by using the traditional algorithm,the computation results indicate that VaRs calculated by GA can cover the market risk,basing on whether linear model GARCH(1,1) or non-linear model EGARCH(1,1),and well perform over the conventional algorithm on the aspect of the accuracy.
Keywords:GA  BHHH  VaR  GARCH  EGARCH
本文献已被 CNKI 维普 万方数据 等数据库收录!
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