Asymptotically optimal nonparametric one- and two-way analysis of variance tests for the log linear model under censoring

T.J. Terpstra

doi:10.1007/BF02614080

Asymptotically optimal nonparametric one- and two-way analysis of variance tests for the log linear model under censoring

T.J. Terpstra^*

^*Corresponding author for this work

University of Twente

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

20 Downloads (Pure)

Abstract

We consider r ×c populations with failure rates λ_ij(t) satisfying the condition Independent random samples of objects are taken from the populations and their life-times are observed during [0, 1]; in this interval censoring may occur. Asymptotically independent maximin most powerful nonparametric tests are obtained for testing against two-way analysis of variance contiguity alternatives. Analogue results follow for one-way analysis of variance problems with respect to the model

Original language	English
Pages (from-to)	63-90
Number of pages	28
Journal	Metrika: International Journal for Theoretical and Applied Statistics
Volume	36
Issue number	1
DOIs	https://doi.org/10.1007/BF02614080
Publication status	Published - Dec 1989

Keywords

Analysis of variance (ANOVA)
Asymptotic optimal similar tests
Censoring
Contiguity
Counting process
Intensity process
Log linear model
Martingale
Non-parametric theory
Partial likelihood
Stochastic integral
Survival analysis

Access to Document

10.1007/BF02614080

Terpstra1989asymptoticallyFinal published version, 578 KB

Cite this

@article{a27105b90e12472585f7aea9ef503e47,

title = "Asymptotically optimal nonparametric one- and two-way analysis of variance tests for the log linear model under censoring",

abstract = "We consider r ×c populations with failure rates λij(t) satisfying the condition Independent random samples of objects are taken from the populations and their life-times are observed during [0, 1]; in this interval censoring may occur. Asymptotically independent maximin most powerful nonparametric tests are obtained for testing against two-way analysis of variance contiguity alternatives. Analogue results follow for one-way analysis of variance problems with respect to the model",

keywords = "Analysis of variance (ANOVA), Asymptotic optimal similar tests, Censoring, Contiguity, Counting process, Intensity process, Log linear model, Martingale, Non-parametric theory, Partial likelihood, Stochastic integral, Survival analysis",

author = "T.J. Terpstra",

year = "1989",

month = dec,

doi = "10.1007/BF02614080",

language = "English",

volume = "36",

pages = "63--90",

journal = "Metrika: International Journal for Theoretical and Applied Statistics",

issn = "0026-1335",

publisher = "Springer",

number = "1",

}

TY - JOUR

T1 - Asymptotically optimal nonparametric one- and two-way analysis of variance tests for the log linear model under censoring

AU - Terpstra, T.J.

PY - 1989/12

Y1 - 1989/12

N2 - We consider r ×c populations with failure rates λij(t) satisfying the condition Independent random samples of objects are taken from the populations and their life-times are observed during [0, 1]; in this interval censoring may occur. Asymptotically independent maximin most powerful nonparametric tests are obtained for testing against two-way analysis of variance contiguity alternatives. Analogue results follow for one-way analysis of variance problems with respect to the model

AB - We consider r ×c populations with failure rates λij(t) satisfying the condition Independent random samples of objects are taken from the populations and their life-times are observed during [0, 1]; in this interval censoring may occur. Asymptotically independent maximin most powerful nonparametric tests are obtained for testing against two-way analysis of variance contiguity alternatives. Analogue results follow for one-way analysis of variance problems with respect to the model

KW - Analysis of variance (ANOVA)

KW - Asymptotic optimal similar tests

KW - Censoring

KW - Contiguity

KW - Counting process

KW - Intensity process

KW - Log linear model

KW - Martingale

KW - Non-parametric theory

KW - Partial likelihood

KW - Stochastic integral

KW - Survival analysis

UR - http://www.scopus.com/inward/record.url?scp=84950352906&partnerID=8YFLogxK

U2 - 10.1007/BF02614080

DO - 10.1007/BF02614080

M3 - Article

AN - SCOPUS:84950352906

SN - 0026-1335

VL - 36

SP - 63

EP - 90

JO - Metrika: International Journal for Theoretical and Applied Statistics

JF - Metrika: International Journal for Theoretical and Applied Statistics

IS - 1

ER -

Asymptotically optimal nonparametric one- and two-way analysis of variance tests for the log linear model under censoring

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this