Comparative study of embedding methods

C. J. Cellucci; A. M. Albano; P. E. Rapp

Comparative study of embedding methods

C. J. Cellucci^*, A. M. Albano, P. E. Rapp

^*Corresponding author for this work

Military and Emergency Medicine

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

91 Scopus citations

Abstract

The global false nearest neighbors method outperformed three other procedures for determining the embedding dimension. Additionally, when used in combination with GFNN, the first minimum of the mutual information function gave a more successful value of the lag than the first zero of the autocorrelation function. However, before generalizing these results inappropriately, other factors should be considered.

Original language	English
Article number	066210
Pages (from-to)	066210/1-066210/13
Journal	Physical Review E
Volume	67
Issue number	6 2
State	Published - Jun 2003

Cite this

@article{62d792c787b045f09b851c21df8a3634,

title = "Comparative study of embedding methods",

abstract = "The global false nearest neighbors method outperformed three other procedures for determining the embedding dimension. Additionally, when used in combination with GFNN, the first minimum of the mutual information function gave a more successful value of the lag than the first zero of the autocorrelation function. However, before generalizing these results inappropriately, other factors should be considered.",

author = "Cellucci, {C. J.} and Albano, {A. M.} and Rapp, {P. E.}",

year = "2003",

month = jun,

language = "English",

volume = "67",

pages = "066210/1--066210/13",

journal = "Physical Review E",

issn = "2470-0045",

publisher = "American Physical Society",

number = "6 2",

}

TY - JOUR

T1 - Comparative study of embedding methods

AU - Cellucci, C. J.

AU - Albano, A. M.

AU - Rapp, P. E.

PY - 2003/6

Y1 - 2003/6

N2 - The global false nearest neighbors method outperformed three other procedures for determining the embedding dimension. Additionally, when used in combination with GFNN, the first minimum of the mutual information function gave a more successful value of the lag than the first zero of the autocorrelation function. However, before generalizing these results inappropriately, other factors should be considered.

AB - The global false nearest neighbors method outperformed three other procedures for determining the embedding dimension. Additionally, when used in combination with GFNN, the first minimum of the mutual information function gave a more successful value of the lag than the first zero of the autocorrelation function. However, before generalizing these results inappropriately, other factors should be considered.

UR - http://www.scopus.com/inward/record.url?scp=42749100939&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:42749100939

SN - 2470-0045

VL - 67

SP - 066210/1-066210/13

JO - Physical Review E

JF - Physical Review E

IS - 6 2

M1 - 066210

ER -