| 基于方差波动多重分形特征的金融时间序列聚类 |
| |
| 引用本文： | 黄超,吴清烈,武忠,朱扬勇.基于方差波动多重分形特征的金融时间序列聚类[J].系统工程,2006,24(6):100-103. |
| |
| 作者姓名： | 黄超 吴清烈 武忠 朱扬勇 |
| |
| 作者单位： | 1. 东南大学,经济管理学院,江苏,南京,210096
2. 复旦大学,计算机与信息技术系,上海,200433 |
| |
| 摘 要： | 提出了一种新的概率函数计算方法，用于研究金融时间序列在方差波动方面的多重分形特征。在此基础上提出了一种基于多重分形的时间序列聚类算法，该算法能够根据不同的分析目的，灵活地使用不同的概率函数以及序列的多重分形特征参量进行聚类。对上海证券市场实际数据的实验结果表明，本文提出的聚类算法是灵活有效的。
|
| 关 键 词： | 多重分形 时间序列 方差波动 聚类 |
| 文章编号： | 1001-4098（2006）06-0100-04 |
| 收稿时间： | 2005-12-10 |
| 修稿时间： | 2005-12-102006-04-27 |
Clustering Financial Time Series Based on Multi-fractal Features of Variance Volatility |
| |
| HUANG Chao,WU Qing-lie,WU Zhong,ZHU Yang-yong.Clustering Financial Time Series Based on Multi-fractal Features of Variance Volatility[J].Systems Engineering,2006,24(6):100-103. |
| |
| Authors: | HUANG Chao WU Qing-lie WU Zhong ZHU Yang-yong |
| |
| Institution: | 1. School of Economics and Management,Southeast University, Nanjing 210096, China ;2. Department of Computing and Information Technology,Fudan University,Shanghai 200433,China |
| |
| Abstract: | A new probabilistic function for studying the multi-fractal features on the volatility of variance of financial time series is proposed. Then a new time series clustering algorithm based on multi-fractal features is brought forward, which can cluster financial time series flexibly according to different purpose by using different probabilistic functions and multifractal features parameters. The experiments conducted on real data of Shanghai securities market show the algorithm is practical and effective. |
| |
| Keywords: | Multi-fractal Features Time Series Variance Volatility Clustering |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
