A comparison of approaches for simultaneous inference of fixed effects for multiple outcomes using linear mixed models

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Standard

A comparison of approaches for simultaneous inference of fixed effects for multiple outcomes using linear mixed models. / Jensen, Signe Marie; Ritz, Christian.

I: Statistics in Medicine, Bind 37, Nr. 16, 2018, s. 2474-2486.

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Harvard

Jensen, SM & Ritz, C 2018, 'A comparison of approaches for simultaneous inference of fixed effects for multiple outcomes using linear mixed models', Statistics in Medicine, bind 37, nr. 16, s. 2474-2486. https://doi.org/10.1002/sim.7666

APA

Jensen, S. M., & Ritz, C. (2018). A comparison of approaches for simultaneous inference of fixed effects for multiple outcomes using linear mixed models. Statistics in Medicine, 37(16), 2474-2486. https://doi.org/10.1002/sim.7666

Vancouver

Jensen SM, Ritz C. A comparison of approaches for simultaneous inference of fixed effects for multiple outcomes using linear mixed models. Statistics in Medicine. 2018;37(16):2474-2486. https://doi.org/10.1002/sim.7666

Author

Jensen, Signe Marie ; Ritz, Christian. / A comparison of approaches for simultaneous inference of fixed effects for multiple outcomes using linear mixed models. I: Statistics in Medicine. 2018 ; Bind 37, Nr. 16. s. 2474-2486.

Bibtex

@article{6418f3c944e64ea8b8f8208c312e31a0,
title = "A comparison of approaches for simultaneous inference of fixed effects for multiple outcomes using linear mixed models",
abstract = "Longitudinal studies with multiple outcomes often pose challenges for the statistical analysis. A joint model including all outcomes has the advantage of incorporating the simultaneous behavior but is often difficult to fit due to computational challenges. We consider 2 alternative approaches to quantify and assess the loss in efficiency as compared with joint modelling when evaluating fixed effects. The first approach is pairwise fitting of pseudolikelihood functions for pairs of outcomes. The second approach recovers correlations between parameter estimates across multiple marginal linear mixed models. The methods are evaluated in terms of a data example both from a study on the effects of milk protein on health in young adolescents and in an extensive simulation study. We find that the 2 alternatives give similar results in settings where an exchangeability condition is met, but otherwise, pairwise fitting shows a larger loss in efficiency than the marginal models approach. Using an alternative to the joint modelling strategy will lead to some but not necessarily a large loss of efficiency for small sample sizes.",
keywords = "Faculty of Science, Correlation, Family-wise error rates, Joint modelling, Marginal models, Multiple testing, Pairwise fitting",
author = "Jensen, {Signe Marie} and Christian Ritz",
note = "CURIS 2018 NEXS 155",
year = "2018",
doi = "10.1002/sim.7666",
language = "English",
volume = "37",
pages = "2474--2486",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "JohnWiley & Sons Ltd",
number = "16",

}

RIS

TY - JOUR

T1 - A comparison of approaches for simultaneous inference of fixed effects for multiple outcomes using linear mixed models

AU - Jensen, Signe Marie

AU - Ritz, Christian

N1 - CURIS 2018 NEXS 155

PY - 2018

Y1 - 2018

N2 - Longitudinal studies with multiple outcomes often pose challenges for the statistical analysis. A joint model including all outcomes has the advantage of incorporating the simultaneous behavior but is often difficult to fit due to computational challenges. We consider 2 alternative approaches to quantify and assess the loss in efficiency as compared with joint modelling when evaluating fixed effects. The first approach is pairwise fitting of pseudolikelihood functions for pairs of outcomes. The second approach recovers correlations between parameter estimates across multiple marginal linear mixed models. The methods are evaluated in terms of a data example both from a study on the effects of milk protein on health in young adolescents and in an extensive simulation study. We find that the 2 alternatives give similar results in settings where an exchangeability condition is met, but otherwise, pairwise fitting shows a larger loss in efficiency than the marginal models approach. Using an alternative to the joint modelling strategy will lead to some but not necessarily a large loss of efficiency for small sample sizes.

AB - Longitudinal studies with multiple outcomes often pose challenges for the statistical analysis. A joint model including all outcomes has the advantage of incorporating the simultaneous behavior but is often difficult to fit due to computational challenges. We consider 2 alternative approaches to quantify and assess the loss in efficiency as compared with joint modelling when evaluating fixed effects. The first approach is pairwise fitting of pseudolikelihood functions for pairs of outcomes. The second approach recovers correlations between parameter estimates across multiple marginal linear mixed models. The methods are evaluated in terms of a data example both from a study on the effects of milk protein on health in young adolescents and in an extensive simulation study. We find that the 2 alternatives give similar results in settings where an exchangeability condition is met, but otherwise, pairwise fitting shows a larger loss in efficiency than the marginal models approach. Using an alternative to the joint modelling strategy will lead to some but not necessarily a large loss of efficiency for small sample sizes.

KW - Faculty of Science

KW - Correlation

KW - Family-wise error rates

KW - Joint modelling

KW - Marginal models

KW - Multiple testing

KW - Pairwise fitting

U2 - 10.1002/sim.7666

DO - 10.1002/sim.7666

M3 - Journal article

C2 - 29664211

VL - 37

SP - 2474

EP - 2486

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 16

ER -

ID: 195554693