Archetypal analysis of spatio-temporal dynamics

Emily Stone; Adele Cutler

doi:10.1016/0167-2789(95)00244-8

Archetypal analysis of spatio-temporal dynamics

Emily Stone, Adele Cutler

Mathematics

Utah State University

Research output: Contribution to journal › Article › peer-review

21 Scopus citations

Abstract

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 language	English
Pages (from-to)	209-224
Number of pages	16
Journal	Physica D: Nonlinear Phenomena
Volume	90
Issue number	3
DOIs	https://doi.org/10.1016/0167-2789(95)00244-8
State	Published - 1996

Funding

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.

Funders	Funder number
Utah State University

Keywords

Archetypal analysis
Dynamical systems
Intermittency
Principal components

Access to Document

10.1016/0167-2789(95)00244-8

Cite this

@article{30a3eb5bfe4544d1886f4119bfb7eeb3,

title = "Archetypal analysis of spatio-temporal dynamics",

abstract = "A comparison is made between the principal component or Karhunen-Lo{\`e}ve 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.",

keywords = "Archetypal analysis, Dynamical systems, Intermittency, Principal components",

author = "Emily Stone and Adele Cutler",

note = "Funding Information: 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.",

year = "1996",

doi = "10.1016/0167-2789(95)00244-8",

language = "English",

volume = "90",

pages = "209--224",

journal = "Physica D: Nonlinear Phenomena",

issn = "0167-2789",

publisher = "Elsevier B.V.",

number = "3",

}

TY - JOUR

T1 - Archetypal analysis of spatio-temporal dynamics

AU - Stone, Emily

AU - Cutler, Adele

N1 - Funding Information: 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.

PY - 1996

Y1 - 1996

N2 - 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.

AB - 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.

KW - Archetypal analysis

KW - Dynamical systems

KW - Intermittency

KW - Principal components

UR - http://www.scopus.com/inward/record.url?scp=0000876180&partnerID=8YFLogxK

U2 - 10.1016/0167-2789(95)00244-8

DO - 10.1016/0167-2789(95)00244-8

M3 - Article

AN - SCOPUS:0000876180

SN - 0167-2789

VL - 90

SP - 209

EP - 224

JO - Physica D: Nonlinear Phenomena

JF - Physica D: Nonlinear Phenomena

IS - 3

ER -

Archetypal analysis of spatio-temporal dynamics

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this