Functional regression analysis using an F test for longitudinal data with large numbers of repeated measures

Xiaowei Yang, Qing Shen, Hongquan Xu, Steven Shoptaw

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Longitudinal data sets from certain fields of biomedical research often consist of several variables repeatedly measured on each subject yielding a large number of observations. This characteristic complicates the use of traditional longitudinal modelling strategies, which were primarily developed for studies with a relatively small number of repeated measures per subject. An innovative way to model such 'wide' data is to apply functional regression analysis, an emerging statistical approach in which observations of the same subject are viewed as a sample from a functional space. Shen and Faraway introduced an F test for linear models with functional responses. This paper illustrates how to apply this F test and functional regression analysis to the setting of longitudinal data. A smoking cessation study for methadone-maintained tobacco smokers is analysed for demonstration. In estimating the treatment effects, the functional regression analysis provides meaningful clinical interpretations, and the functional F test provides consistent results supported by a mixed-effects linear regression model. A simulation study is also conducted under the condition of the smoking data to investigate the statistical power of the F test, Wilks' likelihood ratio test and the linear mixed-effects model using AIC.

Original languageEnglish (US)
Pages (from-to)1552-1566
Number of pages15
JournalStatistics in Medicine
Volume26
Issue number7
DOIs
StatePublished - Mar 30 2007

Fingerprint

F Test
Repeated Measures
Functional Analysis
Longitudinal Data
Regression Analysis
Linear Models
Smoking
Methadone
Smoking Cessation
Linear Mixed Effects Model
Mixed Effects
Statistical Power
Tobacco
Functional Response
Biomedical Research
Treatment Effects
Likelihood Ratio Test
Several Variables
Linear Regression Model
Linear Model

Keywords

  • Functional data analysis
  • Functional F test
  • Functional regression analysis
  • Longitudinal data analysis

ASJC Scopus subject areas

  • Epidemiology

Cite this

Functional regression analysis using an F test for longitudinal data with large numbers of repeated measures. / Yang, Xiaowei; Shen, Qing; Xu, Hongquan; Shoptaw, Steven.

In: Statistics in Medicine, Vol. 26, No. 7, 30.03.2007, p. 1552-1566.

Research output: Contribution to journalArticle

Yang, Xiaowei ; Shen, Qing ; Xu, Hongquan ; Shoptaw, Steven. / Functional regression analysis using an F test for longitudinal data with large numbers of repeated measures. In: Statistics in Medicine. 2007 ; Vol. 26, No. 7. pp. 1552-1566.
@article{cf09811b9511400599adaa1a983dcea7,
title = "Functional regression analysis using an F test for longitudinal data with large numbers of repeated measures",
abstract = "Longitudinal data sets from certain fields of biomedical research often consist of several variables repeatedly measured on each subject yielding a large number of observations. This characteristic complicates the use of traditional longitudinal modelling strategies, which were primarily developed for studies with a relatively small number of repeated measures per subject. An innovative way to model such 'wide' data is to apply functional regression analysis, an emerging statistical approach in which observations of the same subject are viewed as a sample from a functional space. Shen and Faraway introduced an F test for linear models with functional responses. This paper illustrates how to apply this F test and functional regression analysis to the setting of longitudinal data. A smoking cessation study for methadone-maintained tobacco smokers is analysed for demonstration. In estimating the treatment effects, the functional regression analysis provides meaningful clinical interpretations, and the functional F test provides consistent results supported by a mixed-effects linear regression model. A simulation study is also conducted under the condition of the smoking data to investigate the statistical power of the F test, Wilks' likelihood ratio test and the linear mixed-effects model using AIC.",
keywords = "Functional data analysis, Functional F test, Functional regression analysis, Longitudinal data analysis",
author = "Xiaowei Yang and Qing Shen and Hongquan Xu and Steven Shoptaw",
year = "2007",
month = "3",
day = "30",
doi = "10.1002/sim.2609",
language = "English (US)",
volume = "26",
pages = "1552--1566",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "7",

}

TY - JOUR

T1 - Functional regression analysis using an F test for longitudinal data with large numbers of repeated measures

AU - Yang, Xiaowei

AU - Shen, Qing

AU - Xu, Hongquan

AU - Shoptaw, Steven

PY - 2007/3/30

Y1 - 2007/3/30

N2 - Longitudinal data sets from certain fields of biomedical research often consist of several variables repeatedly measured on each subject yielding a large number of observations. This characteristic complicates the use of traditional longitudinal modelling strategies, which were primarily developed for studies with a relatively small number of repeated measures per subject. An innovative way to model such 'wide' data is to apply functional regression analysis, an emerging statistical approach in which observations of the same subject are viewed as a sample from a functional space. Shen and Faraway introduced an F test for linear models with functional responses. This paper illustrates how to apply this F test and functional regression analysis to the setting of longitudinal data. A smoking cessation study for methadone-maintained tobacco smokers is analysed for demonstration. In estimating the treatment effects, the functional regression analysis provides meaningful clinical interpretations, and the functional F test provides consistent results supported by a mixed-effects linear regression model. A simulation study is also conducted under the condition of the smoking data to investigate the statistical power of the F test, Wilks' likelihood ratio test and the linear mixed-effects model using AIC.

AB - Longitudinal data sets from certain fields of biomedical research often consist of several variables repeatedly measured on each subject yielding a large number of observations. This characteristic complicates the use of traditional longitudinal modelling strategies, which were primarily developed for studies with a relatively small number of repeated measures per subject. An innovative way to model such 'wide' data is to apply functional regression analysis, an emerging statistical approach in which observations of the same subject are viewed as a sample from a functional space. Shen and Faraway introduced an F test for linear models with functional responses. This paper illustrates how to apply this F test and functional regression analysis to the setting of longitudinal data. A smoking cessation study for methadone-maintained tobacco smokers is analysed for demonstration. In estimating the treatment effects, the functional regression analysis provides meaningful clinical interpretations, and the functional F test provides consistent results supported by a mixed-effects linear regression model. A simulation study is also conducted under the condition of the smoking data to investigate the statistical power of the F test, Wilks' likelihood ratio test and the linear mixed-effects model using AIC.

KW - Functional data analysis

KW - Functional F test

KW - Functional regression analysis

KW - Longitudinal data analysis

UR - http://www.scopus.com/inward/record.url?scp=33947188066&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947188066&partnerID=8YFLogxK

U2 - 10.1002/sim.2609

DO - 10.1002/sim.2609

M3 - Article

C2 - 16817228

AN - SCOPUS:33947188066

VL - 26

SP - 1552

EP - 1566

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 7

ER -