Plateau onset for correlation dimension: When does it occur?

Mingzhou Ding, Celso Grebogi, Edward Ott, Tim Sauer, J. A. Yorke

Research output: Contribution to journalArticle

149 Citations (Scopus)

Abstract

Chaotic experimental systems are often investigated using delay coordinates. Estimated values of the correlation dimension in delay coordinate space typically increase with the number of delays and eventually reach a plateau (on which the dimension estimate is relatively constant) whose value is commonly taken as an estimate of the correlation dimension D2 of the underlying chaotic attractor. We report a rigorous result which implies that, for long enough data sets, the plateau begins when the number of delay coordinates first exceeds D2. Numerical experiments are presented. We also discuss how lack of sufficient data can produce results that seem to be inconsistent with the theoretical prediction.

Original languageEnglish
Pages (from-to)3872-3875
Number of pages4
JournalPhysical Review Letters
Volume70
Issue number25
DOIs
Publication statusPublished - 21 Jun 1993

Keywords

  • strange attractors
  • systems
  • chaos

Cite this

Plateau onset for correlation dimension : When does it occur? / Ding, Mingzhou; Grebogi, Celso; Ott, Edward ; Sauer, Tim; Yorke, J. A.

In: Physical Review Letters, Vol. 70, No. 25, 21.06.1993, p. 3872-3875.

Research output: Contribution to journalArticle

Ding, Mingzhou ; Grebogi, Celso ; Ott, Edward ; Sauer, Tim ; Yorke, J. A. / Plateau onset for correlation dimension : When does it occur?. In: Physical Review Letters. 1993 ; Vol. 70, No. 25. pp. 3872-3875.
@article{f7cbe0e52ab2435f8d694c17a0380f29,
title = "Plateau onset for correlation dimension: When does it occur?",
abstract = "Chaotic experimental systems are often investigated using delay coordinates. Estimated values of the correlation dimension in delay coordinate space typically increase with the number of delays and eventually reach a plateau (on which the dimension estimate is relatively constant) whose value is commonly taken as an estimate of the correlation dimension D2 of the underlying chaotic attractor. We report a rigorous result which implies that, for long enough data sets, the plateau begins when the number of delay coordinates first exceeds D2. Numerical experiments are presented. We also discuss how lack of sufficient data can produce results that seem to be inconsistent with the theoretical prediction.",
keywords = "strange attractors, systems, chaos",
author = "Mingzhou Ding and Celso Grebogi and Edward Ott and Tim Sauer and Yorke, {J. A.}",
year = "1993",
month = "6",
day = "21",
doi = "10.1103/PhysRevLett.70.3872",
language = "English",
volume = "70",
pages = "3872--3875",
journal = "Physical Review Letters",
issn = "0031-9007",
publisher = "American Physical Society",
number = "25",

}

TY - JOUR

T1 - Plateau onset for correlation dimension

T2 - When does it occur?

AU - Ding, Mingzhou

AU - Grebogi, Celso

AU - Ott, Edward

AU - Sauer, Tim

AU - Yorke, J. A.

PY - 1993/6/21

Y1 - 1993/6/21

N2 - Chaotic experimental systems are often investigated using delay coordinates. Estimated values of the correlation dimension in delay coordinate space typically increase with the number of delays and eventually reach a plateau (on which the dimension estimate is relatively constant) whose value is commonly taken as an estimate of the correlation dimension D2 of the underlying chaotic attractor. We report a rigorous result which implies that, for long enough data sets, the plateau begins when the number of delay coordinates first exceeds D2. Numerical experiments are presented. We also discuss how lack of sufficient data can produce results that seem to be inconsistent with the theoretical prediction.

AB - Chaotic experimental systems are often investigated using delay coordinates. Estimated values of the correlation dimension in delay coordinate space typically increase with the number of delays and eventually reach a plateau (on which the dimension estimate is relatively constant) whose value is commonly taken as an estimate of the correlation dimension D2 of the underlying chaotic attractor. We report a rigorous result which implies that, for long enough data sets, the plateau begins when the number of delay coordinates first exceeds D2. Numerical experiments are presented. We also discuss how lack of sufficient data can produce results that seem to be inconsistent with the theoretical prediction.

KW - strange attractors

KW - systems

KW - chaos

U2 - 10.1103/PhysRevLett.70.3872

DO - 10.1103/PhysRevLett.70.3872

M3 - Article

VL - 70

SP - 3872

EP - 3875

JO - Physical Review Letters

JF - Physical Review Letters

SN - 0031-9007

IS - 25

ER -