Multiple Testing of General Contrasts Using Logical Constraints and Correlations

Peter H. Westfall

doi:10.1080/01621459.1997.10473627

Multiple Testing of General Contrasts Using Logical Constraints and Correlations

Peter H. Westfall

Business Administration

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

199 Scopus citations

Abstract

Use of logical constraints among hypotheses and correlations among test statistics can greatly improve the power of step-down tests. An algorithm for uncovering these logically constrained subsets in a given dataset is described. The multiple testing results are summarized using adjusted p values, which incorporate the relevant dependence structures and logical constraints. These adjusted p values are computed consistently and efficiently using a generalized least squares hybrid of simple and control-variate Monte Carlo methods, and the results are compared to alternative stepwise testing procedures.

Original language	English
Pages (from-to)	299-306
Number of pages	8
Journal	Journal of the American Statistical Association
Volume	92
Issue number	437
DOIs	https://doi.org/10.1080/01621459.1997.10473627
State	Published - Mar 1 1997

Keywords

Adjusted p value
Control variate
Linear model
Monte Carlo
Multiple comparisons
Simultaneous inference

Access to Document

10.1080/01621459.1997.10473627

Cite this

@article{7f79bf19239e4e9a8cc570d123a899e7,

title = "Multiple Testing of General Contrasts Using Logical Constraints and Correlations",

abstract = "Use of logical constraints among hypotheses and correlations among test statistics can greatly improve the power of step-down tests. An algorithm for uncovering these logically constrained subsets in a given dataset is described. The multiple testing results are summarized using adjusted p values, which incorporate the relevant dependence structures and logical constraints. These adjusted p values are computed consistently and efficiently using a generalized least squares hybrid of simple and control-variate Monte Carlo methods, and the results are compared to alternative stepwise testing procedures.",

keywords = "Adjusted p value, Control variate, Linear model, Monte Carlo, Multiple comparisons, Simultaneous inference",

author = "Westfall, {Peter H.}",

year = "1997",

month = mar,

day = "1",

doi = "10.1080/01621459.1997.10473627",

language = "English",

volume = "92",

pages = "299--306",

journal = "Journal of the American Statistical Association",

issn = "0162-1459",

number = "437",

}

TY - JOUR

T1 - Multiple Testing of General Contrasts Using Logical Constraints and Correlations

AU - Westfall, Peter H.

PY - 1997/3/1

Y1 - 1997/3/1

N2 - Use of logical constraints among hypotheses and correlations among test statistics can greatly improve the power of step-down tests. An algorithm for uncovering these logically constrained subsets in a given dataset is described. The multiple testing results are summarized using adjusted p values, which incorporate the relevant dependence structures and logical constraints. These adjusted p values are computed consistently and efficiently using a generalized least squares hybrid of simple and control-variate Monte Carlo methods, and the results are compared to alternative stepwise testing procedures.

AB - Use of logical constraints among hypotheses and correlations among test statistics can greatly improve the power of step-down tests. An algorithm for uncovering these logically constrained subsets in a given dataset is described. The multiple testing results are summarized using adjusted p values, which incorporate the relevant dependence structures and logical constraints. These adjusted p values are computed consistently and efficiently using a generalized least squares hybrid of simple and control-variate Monte Carlo methods, and the results are compared to alternative stepwise testing procedures.

KW - Adjusted p value

KW - Control variate

KW - Linear model

KW - Monte Carlo

KW - Multiple comparisons

KW - Simultaneous inference

UR - http://www.scopus.com/inward/record.url?scp=0031482736&partnerID=8YFLogxK

U2 - 10.1080/01621459.1997.10473627

DO - 10.1080/01621459.1997.10473627

M3 - Article

AN - SCOPUS:0031482736

SN - 0162-1459

VL - 92

SP - 299

EP - 306

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

IS - 437

ER -

Multiple Testing of General Contrasts Using Logical Constraints and Correlations

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this