Heuristic segmentation of a nonstationary time series

Kensuke Fukuda*, H. Eugene Stanley, Luis A.Nunes Amaral

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

36 Scopus citations


The validity of a heuristic segmentation algorithm was studied by analyzing surrogate time series with different statistical properties. The algorithm can split the time series into a set of stationary segments with correct statistical properties, if a given nonstationary time series has stationary periods whose length is distributed as a power law. It also cuts time series into the stationary segments if there are long-range temporal correlations of the fluctuations around the mean of the segment. The results suggest that one must calculate the ratio between the standard deviation of the mean value of the segment and the standard deviation of the fluctuations within a segment after performing the segmentation.

Original languageEnglish (US)
Article number021108
Pages (from-to)021108-1-021108-12
JournalPhysical Review E - Statistical, Nonlinear, and Soft Matter Physics
Issue number2 1
StatePublished - Feb 2004

ASJC Scopus subject areas

  • Condensed Matter Physics
  • Statistical and Nonlinear Physics
  • Statistics and Probability


Dive into the research topics of 'Heuristic segmentation of a nonstationary time series'. Together they form a unique fingerprint.

Cite this