Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (34): 129-131.DOI: 10.3778/j.issn.1002-8331.2008.34.040

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Hierarchical algorithm for time series similar pattern matching based on EMD

GUO Yan-qin,JIA Su-ling   

  1. Department of Economics and Management,Beihang University,Beijing 100083,China
  • Received:2008-05-20 Revised:2008-06-23 Online:2008-12-01 Published:2008-12-01
  • Contact: GUO Yan-qin

基于EMD方法的时间序列分层相似性匹配算法

郭艳琴,贾素玲   

  1. 北京航空航天大学 经济管理学院,北京 100083
  • 通讯作者: 郭艳琴

Abstract: Data mining of time series is an important part of DM(Data Mining).This article first extracts the trend information from financial series using EMD method which is applicable for non-stationary and non-linear time series.Then the hierarchical algorithm for time series similar pattern matching is proposed based on the result of trend extraction which improves the effectiveness of similar pattern matching and decreases the redundancy.

Key words: Empirical Mode Decomposition(EMD), trend extraction, similar pattern matching time series

摘要: 时间序列数据挖掘是时态数据挖掘的一个重要方面,针对金融时间序列非稳定、非线性的特点,使用EMD方法进行序列趋势的提取,得到了原始时间序列的长期趋势。在此基础上提出了子序列分层匹配算法,首先进行时间序列趋势的粗匹配,在结果集中进一步进行细节匹配,与传统方法相比,提高了相似性匹配的效率,减少了结果集的冗余。

关键词: 经验模式分解, 趋势提取, 时间序列相似性匹配