A Unified Approach to Functional Principal Component Analysis and Functional Multiple-Set Canonical Correlation

Ji Yeh Choi, Heungsun Hwang, Michio Yamamoto, Kwanghee Jung, Todd S. Woodward

Research output: Contribution to journalArticle

Abstract

Functional principal component analysis (FPCA) and functional multiple-set canonical correlation analysis (FMCCA) are data reduction techniques for functional data that are collected in the form of smooth curves or functions over a continuum such as time or space. In FPCA, low-dimensional components are extracted from a single functional dataset such that they explain the most variance of the dataset, whereas in FMCCA, low-dimensional components are obtained from each of multiple functional datasets in such a way that the associations among the components are maximized across the different sets. In this paper, we propose a unified approach to FPCA and FMCCA. The proposed approach subsumes both techniques as special cases. Furthermore, it permits a compromise between the techniques, such that components are obtained from each set of functional data to maximize their associations across different datasets, while accounting for the variance of the data well. We propose a single optimization criterion for the proposed approach, and develop an alternating regularized least squares algorithm to minimize the criterion in combination with basis function approximations to functions. We conduct a simulation study to investigate the performance of the proposed approach based on synthetic data. We also apply the approach for the analysis of multiple-subject functional magnetic resonance imaging data to obtain low-dimensional components of blood-oxygen level-dependent signal changes of the brain over time, which are highly correlated across the subjects as well as representative of the data. The extracted components are used to identify networks of neural activity that are commonly activated across the subjects while carrying out a working memory task.

Original languageEnglish
Pages (from-to)427-441
Number of pages15
JournalPsychometrika
Volume82
Issue number2
DOIs
Publication statusPublished - Jun 1 2017
Externally publishedYes

Fingerprint

Functional Principal Component Analysis
Canonical Correlation
Principal Component Analysis
Principal component analysis
Canonical Correlation Analysis
Functional Data
Brain
Data reduction
Blood
Alternating Least Squares
Data storage equipment
Working Memory
Least-Squares Analysis
Oxygen
Short-Term Memory
Functional Magnetic Resonance Imaging
Data Reduction
Least Square Algorithm
Function Approximation
Synthetic Data

Keywords

  • alternating regularized least squares algorithm
  • functional data
  • functional multiple-set canonical correlation analysis
  • functional principal component analysis

ASJC Scopus subject areas

  • Psychology(all)
  • Applied Mathematics

Cite this

A Unified Approach to Functional Principal Component Analysis and Functional Multiple-Set Canonical Correlation. / Choi, Ji Yeh; Hwang, Heungsun; Yamamoto, Michio; Jung, Kwanghee; Woodward, Todd S.

In: Psychometrika, Vol. 82, No. 2, 01.06.2017, p. 427-441.

Research output: Contribution to journalArticle

Choi, Ji Yeh ; Hwang, Heungsun ; Yamamoto, Michio ; Jung, Kwanghee ; Woodward, Todd S. / A Unified Approach to Functional Principal Component Analysis and Functional Multiple-Set Canonical Correlation. In: Psychometrika. 2017 ; Vol. 82, No. 2. pp. 427-441.
@article{c49c3ade4d7042c48ab948ad7e4f7081,
title = "A Unified Approach to Functional Principal Component Analysis and Functional Multiple-Set Canonical Correlation",
abstract = "Functional principal component analysis (FPCA) and functional multiple-set canonical correlation analysis (FMCCA) are data reduction techniques for functional data that are collected in the form of smooth curves or functions over a continuum such as time or space. In FPCA, low-dimensional components are extracted from a single functional dataset such that they explain the most variance of the dataset, whereas in FMCCA, low-dimensional components are obtained from each of multiple functional datasets in such a way that the associations among the components are maximized across the different sets. In this paper, we propose a unified approach to FPCA and FMCCA. The proposed approach subsumes both techniques as special cases. Furthermore, it permits a compromise between the techniques, such that components are obtained from each set of functional data to maximize their associations across different datasets, while accounting for the variance of the data well. We propose a single optimization criterion for the proposed approach, and develop an alternating regularized least squares algorithm to minimize the criterion in combination with basis function approximations to functions. We conduct a simulation study to investigate the performance of the proposed approach based on synthetic data. We also apply the approach for the analysis of multiple-subject functional magnetic resonance imaging data to obtain low-dimensional components of blood-oxygen level-dependent signal changes of the brain over time, which are highly correlated across the subjects as well as representative of the data. The extracted components are used to identify networks of neural activity that are commonly activated across the subjects while carrying out a working memory task.",
keywords = "alternating regularized least squares algorithm, functional data, functional multiple-set canonical correlation analysis, functional principal component analysis",
author = "Choi, {Ji Yeh} and Heungsun Hwang and Michio Yamamoto and Kwanghee Jung and Woodward, {Todd S.}",
year = "2017",
month = "6",
day = "1",
doi = "10.1007/s11336-015-9478-5",
language = "English",
volume = "82",
pages = "427--441",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "2",

}

TY - JOUR

T1 - A Unified Approach to Functional Principal Component Analysis and Functional Multiple-Set Canonical Correlation

AU - Choi, Ji Yeh

AU - Hwang, Heungsun

AU - Yamamoto, Michio

AU - Jung, Kwanghee

AU - Woodward, Todd S.

PY - 2017/6/1

Y1 - 2017/6/1

N2 - Functional principal component analysis (FPCA) and functional multiple-set canonical correlation analysis (FMCCA) are data reduction techniques for functional data that are collected in the form of smooth curves or functions over a continuum such as time or space. In FPCA, low-dimensional components are extracted from a single functional dataset such that they explain the most variance of the dataset, whereas in FMCCA, low-dimensional components are obtained from each of multiple functional datasets in such a way that the associations among the components are maximized across the different sets. In this paper, we propose a unified approach to FPCA and FMCCA. The proposed approach subsumes both techniques as special cases. Furthermore, it permits a compromise between the techniques, such that components are obtained from each set of functional data to maximize their associations across different datasets, while accounting for the variance of the data well. We propose a single optimization criterion for the proposed approach, and develop an alternating regularized least squares algorithm to minimize the criterion in combination with basis function approximations to functions. We conduct a simulation study to investigate the performance of the proposed approach based on synthetic data. We also apply the approach for the analysis of multiple-subject functional magnetic resonance imaging data to obtain low-dimensional components of blood-oxygen level-dependent signal changes of the brain over time, which are highly correlated across the subjects as well as representative of the data. The extracted components are used to identify networks of neural activity that are commonly activated across the subjects while carrying out a working memory task.

AB - Functional principal component analysis (FPCA) and functional multiple-set canonical correlation analysis (FMCCA) are data reduction techniques for functional data that are collected in the form of smooth curves or functions over a continuum such as time or space. In FPCA, low-dimensional components are extracted from a single functional dataset such that they explain the most variance of the dataset, whereas in FMCCA, low-dimensional components are obtained from each of multiple functional datasets in such a way that the associations among the components are maximized across the different sets. In this paper, we propose a unified approach to FPCA and FMCCA. The proposed approach subsumes both techniques as special cases. Furthermore, it permits a compromise between the techniques, such that components are obtained from each set of functional data to maximize their associations across different datasets, while accounting for the variance of the data well. We propose a single optimization criterion for the proposed approach, and develop an alternating regularized least squares algorithm to minimize the criterion in combination with basis function approximations to functions. We conduct a simulation study to investigate the performance of the proposed approach based on synthetic data. We also apply the approach for the analysis of multiple-subject functional magnetic resonance imaging data to obtain low-dimensional components of blood-oxygen level-dependent signal changes of the brain over time, which are highly correlated across the subjects as well as representative of the data. The extracted components are used to identify networks of neural activity that are commonly activated across the subjects while carrying out a working memory task.

KW - alternating regularized least squares algorithm

KW - functional data

KW - functional multiple-set canonical correlation analysis

KW - functional principal component analysis

UR - http://www.scopus.com/inward/record.url?scp=84957679894&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957679894&partnerID=8YFLogxK

U2 - 10.1007/s11336-015-9478-5

DO - 10.1007/s11336-015-9478-5

M3 - Article

C2 - 26856725

AN - SCOPUS:84957679894

VL - 82

SP - 427

EP - 441

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 2

ER -