File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Test de différence de contrastes et somme pondérée de khi-deux

TitleTest de différence de contrastes et somme pondérée de khi-deux
Authors
KeywordsCoding
Conditional Pseudolikelihood
Contrast-Difference Test
Inhomogeneous Markov Chain
Minimum-Contrast Estimation
Whittle's Contrast
Issue Date1996
PublisherStatistical Society of Canada. The Journal's web site is located at http://www.mat.ulaval.ca/cjs
Citation
Canadian Journal Of Statistics, 1996, v. 24 n. 1, p. 115-130 How to Cite?
AbstractAfter recalling the framework of minimum-contrast estimation, its consistency and its asymptotic normality, we highlight the fact that these results do not require any stationary or ergodicity assumptions. The asymptotic distribution of the underlying contrast difference test is a weighted sum of independent chi-square variables having one degree of freedom each. We illustrate these results in three contexts: (1) a nonhomogeneous Markov chain with likelihood contrast; (2) a Markov field with coding, pseudolikelihood or likelihood contrasts; (3) a not necessarily Gaussian time series with Whittle's contrast. In contexts (2) and (3), we compare experimentally the power of the likelihood-ratio test with those of the other contrast-difference tests.
Persistent Identifierhttp://hdl.handle.net/10722/172374
ISSN
2015 Impact Factor: 0.413
2015 SCImago Journal Rankings: 0.737
References

 

DC FieldValueLanguage
dc.contributor.authorBayomog, Sen_US
dc.contributor.authorGuyon, Xen_US
dc.contributor.authorHardouin, Cen_US
dc.contributor.authorJianfeng, Yen_US
dc.date.accessioned2012-10-30T06:22:11Z-
dc.date.available2012-10-30T06:22:11Z-
dc.date.issued1996en_US
dc.identifier.citationCanadian Journal Of Statistics, 1996, v. 24 n. 1, p. 115-130en_US
dc.identifier.issn0319-5724en_US
dc.identifier.urihttp://hdl.handle.net/10722/172374-
dc.description.abstractAfter recalling the framework of minimum-contrast estimation, its consistency and its asymptotic normality, we highlight the fact that these results do not require any stationary or ergodicity assumptions. The asymptotic distribution of the underlying contrast difference test is a weighted sum of independent chi-square variables having one degree of freedom each. We illustrate these results in three contexts: (1) a nonhomogeneous Markov chain with likelihood contrast; (2) a Markov field with coding, pseudolikelihood or likelihood contrasts; (3) a not necessarily Gaussian time series with Whittle's contrast. In contexts (2) and (3), we compare experimentally the power of the likelihood-ratio test with those of the other contrast-difference tests.en_US
dc.languageengen_US
dc.publisherStatistical Society of Canada. The Journal's web site is located at http://www.mat.ulaval.ca/cjsen_US
dc.relation.ispartofCanadian Journal of Statisticsen_US
dc.subjectCodingen_US
dc.subjectConditional Pseudolikelihooden_US
dc.subjectContrast-Difference Testen_US
dc.subjectInhomogeneous Markov Chainen_US
dc.subjectMinimum-Contrast Estimationen_US
dc.subjectWhittle's Contrasten_US
dc.titleTest de différence de contrastes et somme pondérée de khi-deuxen_US
dc.typeArticleen_US
dc.identifier.emailJianfeng, Y: jeffyao@hku.hken_US
dc.identifier.authorityJianfeng, Y=rp01473en_US
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.scopuseid_2-s2.0-0030530918en_US
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-0030530918&selection=ref&src=s&origin=recordpageen_US
dc.identifier.volume24en_US
dc.identifier.issue1en_US
dc.identifier.spage115en_US
dc.identifier.epage130en_US
dc.publisher.placeCanadaen_US
dc.identifier.scopusauthoridBayomog, S=6507753880en_US
dc.identifier.scopusauthoridGuyon, X=6602587667en_US
dc.identifier.scopusauthoridHardouin, C=15032906000en_US
dc.identifier.scopusauthoridJianfeng, Y=7403503451en_US

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats