Multivariate many-to-one procedures with applications to preclinical trials

Research output: Contribution to journalArticleResearchpeer review

Authors

  • Siegfried Kropf
  • Ludwig A. Hothorn
  • Jürgen Läuter

Research Organisations

External Research Organisations

  • Otto-von-Guericke University Magdeburg
View graph of relations

Details

Original languageEnglish
Pages (from-to)433-447
Number of pages15
JournalTherapeutic Innovation & Regulatory Science
Volume31
Issue number2
Publication statusPublished - 30 Dec 1997

Abstract

Comparisons of several treatments with a control represent a standard situation in preclinical trials. Usually, they are considered with a single variable, resulting in multiple test procedures such as the Dunnett test (1). Here, the multivariate many-to-one problem is considered, where several variables are observed on each individual of the control and treatment groups. Classical MANOVA tests and their derivatives for the many-to-one problem require large sample sizes in order to be powerful if the dimension is high. In this paper, a new class of stabilized multivariate tests proposed by Läuter (2) and Läuter, Glimm, and Kropf (3) is extended to this special design. The new tests are based on linear scores which are derived in a certain way from the original variables. They utilize factorial relations among the variables. It is shown here that the procedures keep the multiple level. In simulation experiments several versions of multivariate tests are compared with each other. Standard approaches are included as well as different score versions and a comparison of Dunnett-like procedures with Bonferroni-type procedures. Generally, an improved power of the new tests compared to standard procedures is demonstrated.

Keywords

    Dunnett test, Many-to-one procedures, Multivariate tests, Principal component test, Stabilized scores

ASJC Scopus subject areas

Sustainable Development Goals

Cite this

Multivariate many-to-one procedures with applications to preclinical trials. / Kropf, Siegfried; Hothorn, Ludwig A.; Läuter, Jürgen.
In: Therapeutic Innovation & Regulatory Science, Vol. 31, No. 2, 30.12.1997, p. 433-447.

Research output: Contribution to journalArticleResearchpeer review

Kropf S, Hothorn LA, Läuter J. Multivariate many-to-one procedures with applications to preclinical trials. Therapeutic Innovation & Regulatory Science. 1997 Dec 30;31(2):433-447. doi: 10.1177/009286159703100214, 10.15488/3023
Kropf, Siegfried ; Hothorn, Ludwig A. ; Läuter, Jürgen. / Multivariate many-to-one procedures with applications to preclinical trials. In: Therapeutic Innovation & Regulatory Science. 1997 ; Vol. 31, No. 2. pp. 433-447.
Download
@article{c546ad7000b84a3bbfd7890bfe9739c5,
title = "Multivariate many-to-one procedures with applications to preclinical trials",
abstract = "Comparisons of several treatments with a control represent a standard situation in preclinical trials. Usually, they are considered with a single variable, resulting in multiple test procedures such as the Dunnett test (1). Here, the multivariate many-to-one problem is considered, where several variables are observed on each individual of the control and treatment groups. Classical MANOVA tests and their derivatives for the many-to-one problem require large sample sizes in order to be powerful if the dimension is high. In this paper, a new class of stabilized multivariate tests proposed by L{\"a}uter (2) and L{\"a}uter, Glimm, and Kropf (3) is extended to this special design. The new tests are based on linear scores which are derived in a certain way from the original variables. They utilize factorial relations among the variables. It is shown here that the procedures keep the multiple level. In simulation experiments several versions of multivariate tests are compared with each other. Standard approaches are included as well as different score versions and a comparison of Dunnett-like procedures with Bonferroni-type procedures. Generally, an improved power of the new tests compared to standard procedures is demonstrated.",
keywords = "Dunnett test, Many-to-one procedures, Multivariate tests, Principal component test, Stabilized scores",
author = "Siegfried Kropf and Hothorn, {Ludwig A.} and J{\"u}rgen L{\"a}uter",
year = "1997",
month = dec,
day = "30",
doi = "10.1177/009286159703100214",
language = "English",
volume = "31",
pages = "433--447",
journal = "Therapeutic Innovation & Regulatory Science",
issn = "2168-4790",
publisher = "SAGE Publications Inc.",
number = "2",

}

Download

TY - JOUR

T1 - Multivariate many-to-one procedures with applications to preclinical trials

AU - Kropf, Siegfried

AU - Hothorn, Ludwig A.

AU - Läuter, Jürgen

PY - 1997/12/30

Y1 - 1997/12/30

N2 - Comparisons of several treatments with a control represent a standard situation in preclinical trials. Usually, they are considered with a single variable, resulting in multiple test procedures such as the Dunnett test (1). Here, the multivariate many-to-one problem is considered, where several variables are observed on each individual of the control and treatment groups. Classical MANOVA tests and their derivatives for the many-to-one problem require large sample sizes in order to be powerful if the dimension is high. In this paper, a new class of stabilized multivariate tests proposed by Läuter (2) and Läuter, Glimm, and Kropf (3) is extended to this special design. The new tests are based on linear scores which are derived in a certain way from the original variables. They utilize factorial relations among the variables. It is shown here that the procedures keep the multiple level. In simulation experiments several versions of multivariate tests are compared with each other. Standard approaches are included as well as different score versions and a comparison of Dunnett-like procedures with Bonferroni-type procedures. Generally, an improved power of the new tests compared to standard procedures is demonstrated.

AB - Comparisons of several treatments with a control represent a standard situation in preclinical trials. Usually, they are considered with a single variable, resulting in multiple test procedures such as the Dunnett test (1). Here, the multivariate many-to-one problem is considered, where several variables are observed on each individual of the control and treatment groups. Classical MANOVA tests and their derivatives for the many-to-one problem require large sample sizes in order to be powerful if the dimension is high. In this paper, a new class of stabilized multivariate tests proposed by Läuter (2) and Läuter, Glimm, and Kropf (3) is extended to this special design. The new tests are based on linear scores which are derived in a certain way from the original variables. They utilize factorial relations among the variables. It is shown here that the procedures keep the multiple level. In simulation experiments several versions of multivariate tests are compared with each other. Standard approaches are included as well as different score versions and a comparison of Dunnett-like procedures with Bonferroni-type procedures. Generally, an improved power of the new tests compared to standard procedures is demonstrated.

KW - Dunnett test

KW - Many-to-one procedures

KW - Multivariate tests

KW - Principal component test

KW - Stabilized scores

UR - http://www.scopus.com/inward/record.url?scp=84993740394&partnerID=8YFLogxK

U2 - 10.1177/009286159703100214

DO - 10.1177/009286159703100214

M3 - Article

AN - SCOPUS:84993740394

VL - 31

SP - 433

EP - 447

JO - Therapeutic Innovation & Regulatory Science

JF - Therapeutic Innovation & Regulatory Science

SN - 2168-4790

IS - 2

ER -