Curve registration by local regression

A. Kneip, Xiaochun Li, K. B. MacGibbon, J. O. Ramsay

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

Functional data analysis involves the extension of familiar statistical procedures such as principal-components analysis, linear modelling and canonical correlation analysis to data where the raw observation is a function xi(t). An essential preliminary to a functional data analysis is often the registration or alignment of salient curve features by suitable monotone transformations hi(t). In effect, this conceptualizes variation among functions as being composed of two aspects: phase and amplitude. Registration aims to remove phase variation as a preliminary to statistical analyses of amplitude variation. A local nonlinear regression technique is described for identifying the smooth monotone transformations hi, and is illustrated by analyses of simulated and actual data.

Original languageEnglish (US)
Pages (from-to)19-29
Number of pages11
JournalCanadian Journal of Statistics
Volume28
Issue number1
DOIs
StatePublished - Jan 1 2000
Externally publishedYes

Fingerprint

Local Regression
Registration
Functional Data Analysis
Curve
Monotone
Canonical Correlation Analysis
Nonlinear Regression
Principal Component Analysis
Alignment
Modeling

Keywords

  • Dynamic time warping
  • Monotone functions
  • Time warping

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Curve registration by local regression. / Kneip, A.; Li, Xiaochun; MacGibbon, K. B.; Ramsay, J. O.

In: Canadian Journal of Statistics, Vol. 28, No. 1, 01.01.2000, p. 19-29.

Research output: Contribution to journalArticle

Kneip, A, Li, X, MacGibbon, KB & Ramsay, JO 2000, 'Curve registration by local regression', Canadian Journal of Statistics, vol. 28, no. 1, pp. 19-29. https://doi.org/10.2307/3315251.n
Kneip, A. ; Li, Xiaochun ; MacGibbon, K. B. ; Ramsay, J. O. / Curve registration by local regression. In: Canadian Journal of Statistics. 2000 ; Vol. 28, No. 1. pp. 19-29.
@article{7cd862fbfa304a098838f8de062fec6b,
title = "Curve registration by local regression",
abstract = "Functional data analysis involves the extension of familiar statistical procedures such as principal-components analysis, linear modelling and canonical correlation analysis to data where the raw observation is a function xi(t). An essential preliminary to a functional data analysis is often the registration or alignment of salient curve features by suitable monotone transformations hi(t). In effect, this conceptualizes variation among functions as being composed of two aspects: phase and amplitude. Registration aims to remove phase variation as a preliminary to statistical analyses of amplitude variation. A local nonlinear regression technique is described for identifying the smooth monotone transformations hi, and is illustrated by analyses of simulated and actual data.",
keywords = "Dynamic time warping, Monotone functions, Time warping",
author = "A. Kneip and Xiaochun Li and MacGibbon, {K. B.} and Ramsay, {J. O.}",
year = "2000",
month = "1",
day = "1",
doi = "10.2307/3315251.n",
language = "English (US)",
volume = "28",
pages = "19--29",
journal = "Canadian Journal of Statistics",
issn = "0319-5724",
publisher = "Statistical Society of Canada",
number = "1",

}

TY - JOUR

T1 - Curve registration by local regression

AU - Kneip, A.

AU - Li, Xiaochun

AU - MacGibbon, K. B.

AU - Ramsay, J. O.

PY - 2000/1/1

Y1 - 2000/1/1

N2 - Functional data analysis involves the extension of familiar statistical procedures such as principal-components analysis, linear modelling and canonical correlation analysis to data where the raw observation is a function xi(t). An essential preliminary to a functional data analysis is often the registration or alignment of salient curve features by suitable monotone transformations hi(t). In effect, this conceptualizes variation among functions as being composed of two aspects: phase and amplitude. Registration aims to remove phase variation as a preliminary to statistical analyses of amplitude variation. A local nonlinear regression technique is described for identifying the smooth monotone transformations hi, and is illustrated by analyses of simulated and actual data.

AB - Functional data analysis involves the extension of familiar statistical procedures such as principal-components analysis, linear modelling and canonical correlation analysis to data where the raw observation is a function xi(t). An essential preliminary to a functional data analysis is often the registration or alignment of salient curve features by suitable monotone transformations hi(t). In effect, this conceptualizes variation among functions as being composed of two aspects: phase and amplitude. Registration aims to remove phase variation as a preliminary to statistical analyses of amplitude variation. A local nonlinear regression technique is described for identifying the smooth monotone transformations hi, and is illustrated by analyses of simulated and actual data.

KW - Dynamic time warping

KW - Monotone functions

KW - Time warping

UR - http://www.scopus.com/inward/record.url?scp=0034147493&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034147493&partnerID=8YFLogxK

U2 - 10.2307/3315251.n

DO - 10.2307/3315251.n

M3 - Article

AN - SCOPUS:0034147493

VL - 28

SP - 19

EP - 29

JO - Canadian Journal of Statistics

JF - Canadian Journal of Statistics

SN - 0319-5724

IS - 1

ER -