Inference for low- and high-dimensional multigroup repeated measures designs with unequal covariance matrices

Martin Happ; Solomon W. Harrar; Arne C. Bathke

doi:10.1002/bimj.201500064

Inference for low- and high-dimensional multigroup repeated measures designs with unequal covariance matrices

Martin Happ, Solomon W. Harrar, Arne C. Bathke

Dr. Bing Zhang Department of Statistics

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

6 Scopus citations

Abstract

We propose tests for main and simple treatment effects, time effects, as well as treatment by time interactions in possibly high-dimensional multigroup repeated measures designs. The proposed inference procedures extend the work by Brunner et al. (2012) from two to several treatment groups and remain valid for unbalanced data and under unequal covariance matrices. In addition to showing consistency when sample size and dimension tend to infinity at the same rate, we provide finite sample approximations and evaluate their performance in a simulation study, demonstrating better maintenance of the nominal α-level than the popular Box-Greenhouse–Geisser and Huynh–Feldt methods, and a gain in power for informatively increasing dimension. Application is illustrated using electroencephalography (EEG) data from a neurological study involving patients with Alzheimer's disease and other cognitive impairments.

Original language	English
Pages (from-to)	810-830
Number of pages	21
Journal	Biometrical Journal
Volume	58
Issue number	4
DOIs	https://doi.org/10.1002/bimj.201500064
State	Published - Jul 1 2016

Bibliographical note

Funding Information:
We would like to thank Wolfgang Staffen and Yvonne H?ller (Department of Neurology, Christian Doppler Klinik, Paracelsus Medical University, Salzburg, Austria) for providing the EEG data example. Thanks also go to Edgar Brunner (Department of Medical Statistics, University of G?ttingen, Germany) and the anonymous reviewers who have made very helpful suggestions. The theoretical part of the manuscript was mostly developed during a visit of Solomon W. Harrar at Salzburg University, and the host institution, as well as his home institution, University of Kentucky, generously helped in facilitating this visit. This publication was also made possible by a grant from the National Institute of General Medical Sciences (5 U54 GM104944) from the National Institutes of Health.

Publisher Copyright:
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Keywords

Box epsilon
Large dimension
Longitudinal data

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.1002/bimj.201500064

Cite this

@article{541c4f36a8114a48a58aff9b25feb1b9,

title = "Inference for low- and high-dimensional multigroup repeated measures designs with unequal covariance matrices",

abstract = "We propose tests for main and simple treatment effects, time effects, as well as treatment by time interactions in possibly high-dimensional multigroup repeated measures designs. The proposed inference procedures extend the work by Brunner et al. (2012) from two to several treatment groups and remain valid for unbalanced data and under unequal covariance matrices. In addition to showing consistency when sample size and dimension tend to infinity at the same rate, we provide finite sample approximations and evaluate their performance in a simulation study, demonstrating better maintenance of the nominal α-level than the popular Box-Greenhouse–Geisser and Huynh–Feldt methods, and a gain in power for informatively increasing dimension. Application is illustrated using electroencephalography (EEG) data from a neurological study involving patients with Alzheimer's disease and other cognitive impairments.",

keywords = "Box epsilon, Large dimension, Longitudinal data",

author = "Martin Happ and Harrar, {Solomon W.} and Bathke, {Arne C.}",

note = "Funding Information: We would like to thank Wolfgang Staffen and Yvonne H?ller (Department of Neurology, Christian Doppler Klinik, Paracelsus Medical University, Salzburg, Austria) for providing the EEG data example. Thanks also go to Edgar Brunner (Department of Medical Statistics, University of G?ttingen, Germany) and the anonymous reviewers who have made very helpful suggestions. The theoretical part of the manuscript was mostly developed during a visit of Solomon W. Harrar at Salzburg University, and the host institution, as well as his home institution, University of Kentucky, generously helped in facilitating this visit. This publication was also made possible by a grant from the National Institute of General Medical Sciences (5 U54 GM104944) from the National Institutes of Health. Publisher Copyright: {\textcopyright} 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim",

year = "2016",

month = jul,

day = "1",

doi = "10.1002/bimj.201500064",

language = "English",

volume = "58",

pages = "810--830",

number = "4",

}

TY - JOUR

T1 - Inference for low- and high-dimensional multigroup repeated measures designs with unequal covariance matrices

AU - Happ, Martin

AU - Harrar, Solomon W.

AU - Bathke, Arne C.

N1 - Funding Information: We would like to thank Wolfgang Staffen and Yvonne H?ller (Department of Neurology, Christian Doppler Klinik, Paracelsus Medical University, Salzburg, Austria) for providing the EEG data example. Thanks also go to Edgar Brunner (Department of Medical Statistics, University of G?ttingen, Germany) and the anonymous reviewers who have made very helpful suggestions. The theoretical part of the manuscript was mostly developed during a visit of Solomon W. Harrar at Salzburg University, and the host institution, as well as his home institution, University of Kentucky, generously helped in facilitating this visit. This publication was also made possible by a grant from the National Institute of General Medical Sciences (5 U54 GM104944) from the National Institutes of Health. Publisher Copyright: © 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

PY - 2016/7/1

Y1 - 2016/7/1

N2 - We propose tests for main and simple treatment effects, time effects, as well as treatment by time interactions in possibly high-dimensional multigroup repeated measures designs. The proposed inference procedures extend the work by Brunner et al. (2012) from two to several treatment groups and remain valid for unbalanced data and under unequal covariance matrices. In addition to showing consistency when sample size and dimension tend to infinity at the same rate, we provide finite sample approximations and evaluate their performance in a simulation study, demonstrating better maintenance of the nominal α-level than the popular Box-Greenhouse–Geisser and Huynh–Feldt methods, and a gain in power for informatively increasing dimension. Application is illustrated using electroencephalography (EEG) data from a neurological study involving patients with Alzheimer's disease and other cognitive impairments.

AB - We propose tests for main and simple treatment effects, time effects, as well as treatment by time interactions in possibly high-dimensional multigroup repeated measures designs. The proposed inference procedures extend the work by Brunner et al. (2012) from two to several treatment groups and remain valid for unbalanced data and under unequal covariance matrices. In addition to showing consistency when sample size and dimension tend to infinity at the same rate, we provide finite sample approximations and evaluate their performance in a simulation study, demonstrating better maintenance of the nominal α-level than the popular Box-Greenhouse–Geisser and Huynh–Feldt methods, and a gain in power for informatively increasing dimension. Application is illustrated using electroencephalography (EEG) data from a neurological study involving patients with Alzheimer's disease and other cognitive impairments.

KW - Box epsilon

KW - Large dimension

KW - Longitudinal data

UR - http://www.scopus.com/inward/record.url?scp=85027947883&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85027947883&partnerID=8YFLogxK

U2 - 10.1002/bimj.201500064

DO - 10.1002/bimj.201500064

M3 - Article

C2 - 26700536

AN - SCOPUS:85027947883

VL - 58

SP - 810

EP - 830

IS - 4

ER -

Inference for low- and high-dimensional multigroup repeated measures designs with unequal covariance matrices

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this