[LinJLL12]林子雨, 江弋, 赖永炫, 林琛. TPFP:一种新的时间序列延迟相关性分析算法.计算机研究与发展.（EI期刊，已录用，等待发表）.
TPFP:一种新的时间序列延迟相关性分析算法∗
林子雨1+, 江弋1, 赖永炫2, 林琛1
1(厦门大学 计算机科学系, 厦门 361005)
2(厦门大学 软件学院, 厦门 361005)
TPFP:A New Algorithm on Lagged Correlation Analysis between Time Series
LIN Zi-Yu1+, JIANG Yi1, LAI Yong-Xuan2, LIN Chen1
1(Department of Computer Science, Xiamen University, Xiamen 361005, China)
2(School of Software, Xiamen University, Xiamen 361005, China)
+ Corresponding author: E-mail: ziyulin@xmu.edu.cn
Abstract: Three phenomena in lagged correlation analysis between time series, namely, continuous distribution, lag mutation and mutation amplitude distribution feature, are found here based on extensive experiments. It is proved that existing research either has large error when the lag is large, or can not deal with the occasion of lag mutation. Based on the three phenomena, TPFP is proposed here to overcome the disadvantages of the existing methods, which is able to achieve small error when the lag is large, and also it can perform well on the occasion of lag mutation. Extensive experiments show that TPFP can achieve better performance than available methods.
Key words: time series; lagged correlation; correlation analysis; TPFP;
摘 要: 通过实验发现和验证了时间序列延迟相关性分析中存在的三个现象，即连续分布性、延迟突变和突变幅度分布特性；证明了已有研究或者在延迟位置较大时具有较大的误差，或者无法解决延迟突变问题；根据三个实验现象，提出了TPFP方法，它可以克服已有算法的缺陷，在延迟位置较大时也可以具有较小的误差，并且可以有效处理大部分延迟突变情形。大量实验证明，TPFP方法可以比已有方法取得更好的性能。
关键词: 时间序列;延迟相关;相关性分析; 三点预测探查法;