Regularized Multivariate Analysis of Variance

J. Engel; Kas J. Houthuijs; Vasilis Vasiliou; Georgia Charkoftaki

doi:10.1016/B978-0-12-409547-2.14577-9

Regularized Multivariate Analysis of Variance

J. Engel^*, Kas J. Houthuijs^*, Vasilis Vasiliou^*, Georgia Charkoftaki^*

^*Corresponding author for this work

Biometris

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Academic › peer-review

Abstract

Experimental designs in modern experiments in the life sciences often comprise multiple factors. Usually, the number of response variables vastly exceeds the sample size and well-established methods for analysis of multifactor data such as multivariate analysis of variance (MANOVA) cannot be applied. This issue can be circumvented by combining MANOVA with a shrinkage estimator of the correlation matrix. This chapter discusses the principles of this regularized MANOVA (rMANOVA) approach. Two examples from untargeted metabolomics studies are used to demonstrate application of rMANOVA in practice. Focus lies on identification of significant main and interaction effects and the variables associated to them

Original language	English
Title of host publication	Comprehensive Chemometrics
Subtitle of host publication	Chemical and Biochemical Data Analysis
Editors	Steven Brown, Romà Tauler, Beata Walczak
Publisher	Elsevier
Chapter	1.18
Pages	479-494
Volume	1
Edition	2nd
ISBN (Print)	9780444641663
DOIs	https://doi.org/10.1016/B978-0-12-409547-2.14577-9
Publication status	Published - 2 Jun 2020

Access to Document

10.1016/B978-0-12-409547-2.14577-9Licence: Publisher

Cite this

@inbook{6ef4895345e74bbf920e3cb346bee0c4,

title = "Regularized Multivariate Analysis of Variance",

abstract = "Experimental designs in modern experiments in the life sciences often comprise multiple factors. Usually, the number of response variables vastly exceeds the sample size and well-established methods for analysis of multifactor data such as multivariate analysis of variance (MANOVA) cannot be applied. This issue can be circumvented by combining MANOVA with a shrinkage estimator of the correlation matrix. This chapter discusses the principles of this regularized MANOVA (rMANOVA) approach. Two examples from untargeted metabolomics studies are used to demonstrate application of rMANOVA in practice. Focus lies on identification of significant main and interaction effects and the variables associated to them",

author = "J. Engel and Houthuijs, {Kas J.} and Vasilis Vasiliou and Georgia Charkoftaki",

year = "2020",

month = jun,

day = "2",

doi = "10.1016/B978-0-12-409547-2.14577-9",

language = "English",

isbn = "9780444641663",

volume = "1",

pages = "479--494",

editor = "Steven Brown and Rom{\`a} Tauler and Beata Walczak",

booktitle = "Comprehensive Chemometrics",

publisher = "Elsevier",

address = "Netherlands",

edition = "2nd",

}

TY - CHAP

T1 - Regularized Multivariate Analysis of Variance

AU - Engel, J.

AU - Houthuijs, Kas J.

AU - Vasiliou, Vasilis

AU - Charkoftaki, Georgia

PY - 2020/6/2

Y1 - 2020/6/2

N2 - Experimental designs in modern experiments in the life sciences often comprise multiple factors. Usually, the number of response variables vastly exceeds the sample size and well-established methods for analysis of multifactor data such as multivariate analysis of variance (MANOVA) cannot be applied. This issue can be circumvented by combining MANOVA with a shrinkage estimator of the correlation matrix. This chapter discusses the principles of this regularized MANOVA (rMANOVA) approach. Two examples from untargeted metabolomics studies are used to demonstrate application of rMANOVA in practice. Focus lies on identification of significant main and interaction effects and the variables associated to them

AB - Experimental designs in modern experiments in the life sciences often comprise multiple factors. Usually, the number of response variables vastly exceeds the sample size and well-established methods for analysis of multifactor data such as multivariate analysis of variance (MANOVA) cannot be applied. This issue can be circumvented by combining MANOVA with a shrinkage estimator of the correlation matrix. This chapter discusses the principles of this regularized MANOVA (rMANOVA) approach. Two examples from untargeted metabolomics studies are used to demonstrate application of rMANOVA in practice. Focus lies on identification of significant main and interaction effects and the variables associated to them

U2 - 10.1016/B978-0-12-409547-2.14577-9

DO - 10.1016/B978-0-12-409547-2.14577-9

M3 - Chapter

SN - 9780444641663

VL - 1

SP - 479

EP - 494

BT - Comprehensive Chemometrics

A2 - Brown, Steven

A2 - Tauler, Romà

A2 - Walczak, Beata

PB - Elsevier

ER -

Regularized Multivariate Analysis of Variance

Abstract

Access to Document

Fingerprint

Cite this