Simple and accurate one-sided inference from signed roots of likelihood ratios

Thomas J. Diciccio, Michael A. Martin, Steven E. Stern

Research output: Contribution to journalArticleResearchpeer-review

38 Citations (Scopus)

Abstract

The authors propose two methods based on the signed root of the likelihood ratio statistic for one-sided testing of a simple null hypothesis about a scalar parameter in the presence of nuisance parameters. Both methods are third-order accurate and utilise simulation to avoid the need for onerous analytical calculations characteristic of competing saddlepoint procedures. Moreover, the new methods do not require specification of ancillary statistics. The methods respect the conditioning associated with similar tests up to an error of third order, and conditioning on ancillary statistics to an error of second order.

Original languageEnglish
Pages (from-to)67-76
Number of pages10
JournalCanadian Journal of Statistics
Volume29
Issue number1
DOIs
Publication statusPublished - Mar 2001
Externally publishedYes

Fingerprint

Likelihood Ratio
Signed
Ancillary Statistic
Roots
Conditioning
Similar Tests
Likelihood Ratio Statistic
Nuisance Parameter
Saddlepoint
Null hypothesis
Scalar
Specification
Testing
Likelihood ratio
Inference
Simulation
Statistics

Cite this

@article{8858dee3ca294a458dd9445633a2424e,
title = "Simple and accurate one-sided inference from signed roots of likelihood ratios",
abstract = "The authors propose two methods based on the signed root of the likelihood ratio statistic for one-sided testing of a simple null hypothesis about a scalar parameter in the presence of nuisance parameters. Both methods are third-order accurate and utilise simulation to avoid the need for onerous analytical calculations characteristic of competing saddlepoint procedures. Moreover, the new methods do not require specification of ancillary statistics. The methods respect the conditioning associated with similar tests up to an error of third order, and conditioning on ancillary statistics to an error of second order.",
author = "Diciccio, {Thomas J.} and Martin, {Michael A.} and Stern, {Steven E.}",
year = "2001",
month = "3",
doi = "10.2307/3316051",
language = "English",
volume = "29",
pages = "67--76",
journal = "Canadian Journal of Statistics",
issn = "0319-5724",
publisher = "Statistical Society of Canada",
number = "1",

}

Simple and accurate one-sided inference from signed roots of likelihood ratios. / Diciccio, Thomas J.; Martin, Michael A.; Stern, Steven E.

In: Canadian Journal of Statistics, Vol. 29, No. 1, 03.2001, p. 67-76.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Simple and accurate one-sided inference from signed roots of likelihood ratios

AU - Diciccio, Thomas J.

AU - Martin, Michael A.

AU - Stern, Steven E.

PY - 2001/3

Y1 - 2001/3

N2 - The authors propose two methods based on the signed root of the likelihood ratio statistic for one-sided testing of a simple null hypothesis about a scalar parameter in the presence of nuisance parameters. Both methods are third-order accurate and utilise simulation to avoid the need for onerous analytical calculations characteristic of competing saddlepoint procedures. Moreover, the new methods do not require specification of ancillary statistics. The methods respect the conditioning associated with similar tests up to an error of third order, and conditioning on ancillary statistics to an error of second order.

AB - The authors propose two methods based on the signed root of the likelihood ratio statistic for one-sided testing of a simple null hypothesis about a scalar parameter in the presence of nuisance parameters. Both methods are third-order accurate and utilise simulation to avoid the need for onerous analytical calculations characteristic of competing saddlepoint procedures. Moreover, the new methods do not require specification of ancillary statistics. The methods respect the conditioning associated with similar tests up to an error of third order, and conditioning on ancillary statistics to an error of second order.

UR - http://www.scopus.com/inward/record.url?scp=0035617180&partnerID=8YFLogxK

U2 - 10.2307/3316051

DO - 10.2307/3316051

M3 - Article

VL - 29

SP - 67

EP - 76

JO - Canadian Journal of Statistics

JF - Canadian Journal of Statistics

SN - 0319-5724

IS - 1

ER -