Testing with a nuisance parameter present only under the alternative: a score-based approach with application to segmented modelling

Risultato della ricerca: Article

10 Citazioni (Scopus)

Abstract

We introduce a score-type statistic to test for a non-zero regression coefficient when the relevant term involves a nuisance parameter present only under the alternative. Despite the non-regularity and complexity of the problem and unlike the previous approaches, the proposed test statistic does not require the nuisance to be estimated. It is simple to implement by relying on the conventional distributions, such as Normal or t, and it justified in the setting of probabilistic coherence. We focus on testing for the existence of a breakpoint in segmented regression, and illustrate the methodology with an analysis on data of DNA copy number aberrations and gene expression profiles from 97 breast cancer patients; moreover some simulations reveal that the proposed test is more powerful than its competitors previously discussed in literature.
Lingua originaleEnglish
pagine (da-a)3059-3067
Numero di pagine9
RivistaJournal of Statistical Computation and Simulation
Volume86
Stato di pubblicazionePublished - 2016

Fingerprint

Nuisance Parameter
Segmented Regression
Statistics
Testing
Gene Expression Profile
Alternatives
Regression Coefficient
Aberrations
Aberration
Breast Cancer
Modeling
Gene expression
Test Statistic
Statistic
DNA
Methodology
Simulation
Nuisance parameter
Coefficients
Breast cancer

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Applied Mathematics
  • Statistics, Probability and Uncertainty
  • Modelling and Simulation

Cita questo

@article{b5fa07feb71847af801e15ac1a68068d,
title = "Testing with a nuisance parameter present only under the alternative: a score-based approach with application to segmented modelling",
abstract = "We introduce a score-type statistic to test for a non-zero regression coefficient when the relevant term involves a nuisance parameter present only under the alternative. Despite the non-regularity and complexity of the problem and unlike the previous approaches, the proposed test statistic does not require the nuisance to be estimated. It is simple to implement by relying on the conventional distributions, such as Normal or t, and it justified in the setting of probabilistic coherence. We focus on testing for the existence of a breakpoint in segmented regression, and illustrate the methodology with an analysis on data of DNA copy number aberrations and gene expression profiles from 97 breast cancer patients; moreover some simulations reveal that the proposed test is more powerful than its competitors previously discussed in literature.",
keywords = "Breakpoint detection; non-standard inference; score test;piecewise linear; threshold value",
author = "Muggeo, {Vito Michele Rosario}",
year = "2016",
language = "English",
volume = "86",
pages = "3059--3067",
journal = "Journal of Statistical Computation and Simulation",
issn = "0094-9655",
publisher = "Taylor and Francis Ltd.",

}

TY - JOUR

T1 - Testing with a nuisance parameter present only under the alternative: a score-based approach with application to segmented modelling

AU - Muggeo, Vito Michele Rosario

PY - 2016

Y1 - 2016

N2 - We introduce a score-type statistic to test for a non-zero regression coefficient when the relevant term involves a nuisance parameter present only under the alternative. Despite the non-regularity and complexity of the problem and unlike the previous approaches, the proposed test statistic does not require the nuisance to be estimated. It is simple to implement by relying on the conventional distributions, such as Normal or t, and it justified in the setting of probabilistic coherence. We focus on testing for the existence of a breakpoint in segmented regression, and illustrate the methodology with an analysis on data of DNA copy number aberrations and gene expression profiles from 97 breast cancer patients; moreover some simulations reveal that the proposed test is more powerful than its competitors previously discussed in literature.

AB - We introduce a score-type statistic to test for a non-zero regression coefficient when the relevant term involves a nuisance parameter present only under the alternative. Despite the non-regularity and complexity of the problem and unlike the previous approaches, the proposed test statistic does not require the nuisance to be estimated. It is simple to implement by relying on the conventional distributions, such as Normal or t, and it justified in the setting of probabilistic coherence. We focus on testing for the existence of a breakpoint in segmented regression, and illustrate the methodology with an analysis on data of DNA copy number aberrations and gene expression profiles from 97 breast cancer patients; moreover some simulations reveal that the proposed test is more powerful than its competitors previously discussed in literature.

KW - Breakpoint detection; non-standard inference; score test;piecewise linear; threshold value

UR - http://hdl.handle.net/10447/172283

M3 - Article

VL - 86

SP - 3059

EP - 3067

JO - Journal of Statistical Computation and Simulation

JF - Journal of Statistical Computation and Simulation

SN - 0094-9655

ER -