| 基于小波和多重分形的金融时间序列聚类 |
| |
| 引用本文： | 钟维年,高清维,陈燕玲.基于小波和多重分形的金融时间序列聚类[J].系统工程,2009,27(3). |
| |
| 作者姓名： | 钟维年 高清维 陈燕玲 |
| |
| 作者单位： | 钟维年,高清维,ZHONG Wei-nian,GAO Qing-wei(安徽大学,电子科学与技术学院,安徽,合肥,230039);陈燕玲,CHENG Yan-ling(安徽大学,经济学院,安徽,合肥,230039)
|
| |
| 摘 要： | 为研究时间序列波动特征及其聚类,定义了反映序列波动特征的概率测度并考虑长期趋势项对时间序列多重分形分析的影响,采用小波分解与重构的方法去除低阶项,得到消除趋势项后的序列.比较消除趋势项前后序列的多重分形谱,结果显示多重分形谱的形状在消除趋势项后更加趋于钟形.由于多重分形谱可以定量表征序列的波动特征,将此方法应用于上海市场股票价格序列的波动特征分析,结果显示,消除趋势项后的上海市场股票价格序列的多重分形谱用于其波动特征的聚类具有更合理的物理意义.
|
| 关 键 词： | 序列 波动 趋势项 多重分形 股票价格 |
The Clustering of Financial Time Serials Based On Wavelet and Multifractal |
| |
| ZHONG Wei-nian,GAO Qing-wei,CHENG Yan-ling.The Clustering of Financial Time Serials Based On Wavelet and Multifractal[J].Systems Engineering,2009,27(3). |
| |
| Authors: | ZHONG Wei-nian GAO Qing-wei CHENG Yan-ling |
| |
| Institution: | ZHONG Wei-nian1,GAO Qing-wei1,CHENG Yan-ling2(1.School of Electronic Science and Technology,Anhui University,Hefei 230039,China,2.School of Financial,China) |
| |
| Abstract: | To study the fluctuating character and the clustering of financial time serials,we define the probability measure of the fluctuating character of the serials and study the effect of long-range trend item on multifractal analysis.Detrended serials are obtained after eliminating low order terms with the method of wavelet decomposition and restructure.By comparing the multifractal spectrum of the original serials with that of detrended serials,we find the latter's shape is more like bell.Since multifractal spe... |
| |
| Keywords: | Serials Fluctuate Trend Item Multifractal Stock Price |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
