RESEARCH ON THE MODELING OF ARFIMA FOR CHINESE STOCK INDEX BASED ON GENETIC ALGORITHM

摘要：

In this paper we study the long-term memory of Chinese stock index by the Autoregression Fractionally In tegrated Moving Average (ARFIMA) model. The model identification is the crucial stage in building ARFIMA models to overcome the local optima. Thus we employ the genetic algorithms (GA) to do it. The GA-based method obtains the ARFIMA models of the stock index representing the Chinese stock market effectively. The results show that our method is superior to the traditional modeling technologies on the model identification.

关键词： Genetic algorithm Long-term memory ARFIMA Stock index

作者: Haijun Yang Tingting Liu

作者单位: School of Economics and Management, BeiHang University, Beijing 100083, P.R.China

会议类型: 国际会议

会议名称: The Ninth International Conference on Industrial Management(第九届工业管理国际会议 ICIM2008)

会议地点: 日本大阪

会议语种:英文

页码: 656-662

在线出版日期: 2008-09-16（万方平台首次上网日期，不代表论文的发表时间）