Curve registration

J. O. Ramsay, Xiaochun Li

Research output: Contribution to journalArticle

215 Citations (Scopus)

Abstract

Summary. 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 xi is a function. An essential preliminary to a functional data analysis is often the registration or alignment of salient curve features by suitable monotone transformations hi, of the argument t, so that the actual analyses are carried out on the values x,{/7,(f)}- This is referred to as dynamic time warping in the engineering literature. In effect, this conceptualizes variation among functions as being composed of two aspects: horizontal and vertical, or domain and range. A nonparametric function estimation technique is described for identifying the smooth monotone transformations ht and is illustrated by data analyses. A second-order linear stochastic differential equation is proposed to model these components of variation.

Original languageEnglish (US)
Pages (from-to)351-363
Number of pages13
JournalJournal of the Royal Statistical Society. Series B: Statistical Methodology
Volume60
Issue number2
DOIs
StatePublished - Jan 1 1998
Externally publishedYes

Fingerprint

Functional Data Analysis
Registration
Monotone
Dynamic Time Warping
Canonical Correlation Analysis
Curve
Function Estimation
Nonparametric Estimation
Component Model
Principal Component Analysis
Stochastic Equations
Alignment
Horizontal
Vertical
Differential equation
Engineering
Modeling
Range of data
Observation
Warping

Keywords

  • Dynamic time warping
  • Geometric brownian motion
  • Monotone functions
  • Spline
  • Stochastic time
  • Time warping

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Curve registration. / Ramsay, J. O.; Li, Xiaochun.

In: Journal of the Royal Statistical Society. Series B: Statistical Methodology, Vol. 60, No. 2, 01.01.1998, p. 351-363.

Research output: Contribution to journalArticle

@article{b2b2c0553f4b4849a6ceb5ed20f9d5e4,
title = "Curve registration",
abstract = "Summary. 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 xi is a function. An essential preliminary to a functional data analysis is often the registration or alignment of salient curve features by suitable monotone transformations hi, of the argument t, so that the actual analyses are carried out on the values x,{/7,(f)}- This is referred to as dynamic time warping in the engineering literature. In effect, this conceptualizes variation among functions as being composed of two aspects: horizontal and vertical, or domain and range. A nonparametric function estimation technique is described for identifying the smooth monotone transformations ht and is illustrated by data analyses. A second-order linear stochastic differential equation is proposed to model these components of variation.",
keywords = "Dynamic time warping, Geometric brownian motion, Monotone functions, Spline, Stochastic time, Time warping",
author = "Ramsay, {J. O.} and Xiaochun Li",
year = "1998",
month = "1",
day = "1",
doi = "10.1111/1467-9868.00129",
language = "English (US)",
volume = "60",
pages = "351--363",
journal = "Journal of the Royal Statistical Society. Series B: Statistical Methodology",
issn = "1369-7412",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Curve registration

AU - Ramsay, J. O.

AU - Li, Xiaochun

PY - 1998/1/1

Y1 - 1998/1/1

N2 - Summary. 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 xi is a function. An essential preliminary to a functional data analysis is often the registration or alignment of salient curve features by suitable monotone transformations hi, of the argument t, so that the actual analyses are carried out on the values x,{/7,(f)}- This is referred to as dynamic time warping in the engineering literature. In effect, this conceptualizes variation among functions as being composed of two aspects: horizontal and vertical, or domain and range. A nonparametric function estimation technique is described for identifying the smooth monotone transformations ht and is illustrated by data analyses. A second-order linear stochastic differential equation is proposed to model these components of variation.

AB - Summary. 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 xi is a function. An essential preliminary to a functional data analysis is often the registration or alignment of salient curve features by suitable monotone transformations hi, of the argument t, so that the actual analyses are carried out on the values x,{/7,(f)}- This is referred to as dynamic time warping in the engineering literature. In effect, this conceptualizes variation among functions as being composed of two aspects: horizontal and vertical, or domain and range. A nonparametric function estimation technique is described for identifying the smooth monotone transformations ht and is illustrated by data analyses. A second-order linear stochastic differential equation is proposed to model these components of variation.

KW - Dynamic time warping

KW - Geometric brownian motion

KW - Monotone functions

KW - Spline

KW - Stochastic time

KW - Time warping

UR - http://www.scopus.com/inward/record.url?scp=0000965844&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0000965844&partnerID=8YFLogxK

U2 - 10.1111/1467-9868.00129

DO - 10.1111/1467-9868.00129

M3 - Article

VL - 60

SP - 351

EP - 363

JO - Journal of the Royal Statistical Society. Series B: Statistical Methodology

JF - Journal of the Royal Statistical Society. Series B: Statistical Methodology

SN - 1369-7412

IS - 2

ER -