The Tukey trend test: Multiplicity adjustment using multiple marginal models

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

The Tukey trend test: Multiplicity adjustment using multiple marginal models. / Schaarschmidt, Frank; Ritz, Christian; Hothorn, Ludwig A.

In: Biometrics, Vol. 78, No. 2, 2022, p. 789-797.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Schaarschmidt, F, Ritz, C & Hothorn, LA 2022, 'The Tukey trend test: Multiplicity adjustment using multiple marginal models', Biometrics, vol. 78, no. 2, pp. 789-797. https://doi.org/10.1111/biom.13442

APA

Schaarschmidt, F., Ritz, C., & Hothorn, L. A. (2022). The Tukey trend test: Multiplicity adjustment using multiple marginal models. Biometrics, 78(2), 789-797. https://doi.org/10.1111/biom.13442

Vancouver

Schaarschmidt F, Ritz C, Hothorn LA. The Tukey trend test: Multiplicity adjustment using multiple marginal models. Biometrics. 2022;78(2):789-797. https://doi.org/10.1111/biom.13442

Author

Schaarschmidt, Frank ; Ritz, Christian ; Hothorn, Ludwig A. / The Tukey trend test: Multiplicity adjustment using multiple marginal models. In: Biometrics. 2022 ; Vol. 78, No. 2. pp. 789-797.

Bibtex

@article{02715f8a3ba44ccfb6002a30649acb9b,
title = "The Tukey trend test: Multiplicity adjustment using multiple marginal models",
abstract = "In dose-response analysis, it is a challenge to choose appropriate linear or curvilinear shapes when considering multiple, differently scaled endpoints. It has been proposed to fit several marginal regression models that try sets of different transformations of the dose levels as explanatory variables for each endpoint. However, the multiple testing problem underlying this approach, involving correlated parameter estimates for the dose effect between and within endpoints, could only be adjusted heuristically. An asymptotic correction for multiple testing can be derived from the score functions of the marginal regression models. Based on a multivariate t-distribution, the correction provides a one-step adjustment of p-values that accounts for the correlation between estimates from different marginal models. The advantages of the proposed methodology is demonstrated through three example data sets, involving generalized linear models with differently scaled endpoints, differing covariates and a mixed effect model and through simulation results. The methodology is implemented in an R package.",
keywords = "Faculty of Science, Adjustment of p-values, Dose-response, Multiple endpoints, Multivariate normal, Toxicology",
author = "Frank Schaarschmidt and Christian Ritz and Hothorn, {Ludwig A}",
note = "This article is protected by copyright. All rights reserved.",
year = "2022",
doi = "10.1111/biom.13442",
language = "English",
volume = "78",
pages = "789--797",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "2",

}

RIS

TY - JOUR

T1 - The Tukey trend test: Multiplicity adjustment using multiple marginal models

AU - Schaarschmidt, Frank

AU - Ritz, Christian

AU - Hothorn, Ludwig A

N1 - This article is protected by copyright. All rights reserved.

PY - 2022

Y1 - 2022

N2 - In dose-response analysis, it is a challenge to choose appropriate linear or curvilinear shapes when considering multiple, differently scaled endpoints. It has been proposed to fit several marginal regression models that try sets of different transformations of the dose levels as explanatory variables for each endpoint. However, the multiple testing problem underlying this approach, involving correlated parameter estimates for the dose effect between and within endpoints, could only be adjusted heuristically. An asymptotic correction for multiple testing can be derived from the score functions of the marginal regression models. Based on a multivariate t-distribution, the correction provides a one-step adjustment of p-values that accounts for the correlation between estimates from different marginal models. The advantages of the proposed methodology is demonstrated through three example data sets, involving generalized linear models with differently scaled endpoints, differing covariates and a mixed effect model and through simulation results. The methodology is implemented in an R package.

AB - In dose-response analysis, it is a challenge to choose appropriate linear or curvilinear shapes when considering multiple, differently scaled endpoints. It has been proposed to fit several marginal regression models that try sets of different transformations of the dose levels as explanatory variables for each endpoint. However, the multiple testing problem underlying this approach, involving correlated parameter estimates for the dose effect between and within endpoints, could only be adjusted heuristically. An asymptotic correction for multiple testing can be derived from the score functions of the marginal regression models. Based on a multivariate t-distribution, the correction provides a one-step adjustment of p-values that accounts for the correlation between estimates from different marginal models. The advantages of the proposed methodology is demonstrated through three example data sets, involving generalized linear models with differently scaled endpoints, differing covariates and a mixed effect model and through simulation results. The methodology is implemented in an R package.

KW - Faculty of Science

KW - Adjustment of p-values

KW - Dose-response

KW - Multiple endpoints

KW - Multivariate normal

KW - Toxicology

U2 - 10.1111/biom.13442

DO - 10.1111/biom.13442

M3 - Journal article

C2 - 33559878

VL - 78

SP - 789

EP - 797

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 2

ER -

ID: 256626097