Archetypal analysis of spatio-temporal dynamics

Emily Stone, Adele Cutler

Research output: Contribution to journalArticlepeer-review

21 Scopus citations


A comparison is made between the principal component or Karhunen-Loève decomposition of two sets of spatio-temporal data (one numerical, the other experimental) and a new procedure called archetypal analysis (Cutler and Breiman, 1994). Archetypes characterize the convex hull of the data set and the data set can be reconstructed in terms of these values. Archetypes may be more appropriate than KL when the data do not have elliptical distributions, and are often well-suited to studying regimes in which the system spends a long time near a "steady" state, punctuated with quick excursions to outliers in the data set, which may represent intermittency. Other advantages and disadvantages of each method are discussed.

Original languageEnglish
Pages (from-to)209-224
Number of pages16
JournalPhysica D: Nonlinear Phenomena
Issue number3
StatePublished - 1996


One of the authors (E.S.) gratefully acknowledges the financial support of a Utah State University faculty research grant. We would also like to thank Dieter Armbruster, Michael Kirby and Eric Kostelich for valuable discussions and suggestions.

FundersFunder number
Utah State University


    • Archetypal analysis
    • Dynamical systems
    • Intermittency
    • Principal components


    Dive into the research topics of 'Archetypal analysis of spatio-temporal dynamics'. Together they form a unique fingerprint.

    Cite this