Computation of the p-value of the maximum of score tests in the generalized linear model; application to multiple coding

Benoit Liquet; Daniel Commenges

doi:10.1016/j.spl.2004.10.019

Computation of the p-value of the maximum of score tests in the generalized linear model; application to multiple coding

Benoit Liquet^*, Daniel Commenges

^*Corresponding author for this work

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

5 Citations (Scopus)

Abstract

We propose a method to correct the significance level for a series of tests corresponding to several transformations of an explanatory variable in generalized linear model. Correlation between score tests are derived to apply the proposed method.

Original language	English
Pages (from-to)	33-38
Number of pages	6
Journal	Statistics and Probability Letters
Volume	71
Issue number	1
DOIs	https://doi.org/10.1016/j.spl.2004.10.019
Publication status	Published - Jan 2005
Externally published	Yes

Keywords

Generalized linear model
Multiplicity
p-value
Score test

Access to Document

10.1016/j.spl.2004.10.019

Cite this

@article{53cbc8a84a8a4c2ea19e17a1ff59ec6b,

title = "Computation of the p-value of the maximum of score tests in the generalized linear model; application to multiple coding",

abstract = "We propose a method to correct the significance level for a series of tests corresponding to several transformations of an explanatory variable in generalized linear model. Correlation between score tests are derived to apply the proposed method.",

keywords = "Generalized linear model, Multiplicity, p-value, Score test",

author = "Benoit Liquet and Daniel Commenges",

year = "2005",

month = jan,

doi = "10.1016/j.spl.2004.10.019",

language = "English",

volume = "71",

pages = "33--38",

journal = "Statistics and Probability Letters",

issn = "0167-7152",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - Computation of the p-value of the maximum of score tests in the generalized linear model; application to multiple coding

AU - Liquet, Benoit

AU - Commenges, Daniel

PY - 2005/1

Y1 - 2005/1

N2 - We propose a method to correct the significance level for a series of tests corresponding to several transformations of an explanatory variable in generalized linear model. Correlation between score tests are derived to apply the proposed method.

AB - We propose a method to correct the significance level for a series of tests corresponding to several transformations of an explanatory variable in generalized linear model. Correlation between score tests are derived to apply the proposed method.

KW - Generalized linear model

KW - Multiplicity

KW - p-value

KW - Score test

UR - http://www.scopus.com/inward/record.url?scp=27444447097&partnerID=8YFLogxK

U2 - 10.1016/j.spl.2004.10.019

DO - 10.1016/j.spl.2004.10.019

M3 - Article

AN - SCOPUS:27444447097

SN - 0167-7152

VL - 71

SP - 33

EP - 38

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

IS - 1

ER -

Computation of the p-value of the maximum of score tests in the generalized linear model; application to multiple coding

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this