Using a non-homogeneous Poisson model with spatial anisotropy and change-points to study air pollution data

Página do item simplificado
dc.contributor.authorRodrigues, Eliane R.
dc.contributor.authorNicholls, Geoff
dc.contributor.authorTarumoto, Mario H. [UNESP]
dc.contributor.authorTzintzun, Guadalupe
dc.contributor.institutionUniv Nacl Autonoma Mexico
dc.contributor.institutionUniv Oxford
dc.contributor.institutionUniversidade Estadual Paulista (Unesp)
dc.contributor.institutionSecretaria Medio Ambiente & Recursos Nat
dc.date.accessioned2019-10-04T12:38:48Z
dc.date.available2019-10-04T12:38:48Z
dc.date.issued2019-06-01
dc.description.abstractA non-homogeneous Poisson process is used to study the rate at which a pollutant's concentration exceeds a given threshold of interest. An anisotropic spatial model is imposed on the parameters of the Poisson intensity function. The main contribution here is to allow the presence of change-points in time since the data may behave differently for different time frames in a given observational period. Additionally, spatial anisotropy is also imposed on the vector of change-points in order to account for the possible correlation between different sites. Estimation of the parameters of the model is performed using Bayesian inference via Markov chain Monte Carlo algorithms, in particular, Gibbs sampling and Metropolis-Hastings. The different versions of the model are applied to ozone data from the monitoring network of Mexico City, Mexico. An analysis of the results obtained is also given.en
dc.description.affiliationUniv Nacl Autonoma Mexico, Inst Matemat, Area Invest Cient, Mexico City 04510, DF, Mexico
dc.description.affiliationUniv Oxford, Dept Stat, Oxford, England
dc.description.affiliationUniv Estadual Paulista, Dept Estat, Fac Ciencias & Tecnol, Presidente Prudente, SP, Brazil
dc.description.affiliationSecretaria Medio Ambiente & Recursos Nat, Inst Nacl Ecol & Cambio Climat, Mexico City, DF, Mexico
dc.description.affiliationUnespUniv Estadual Paulista, Dept Estat, Fac Ciencias & Tecnol, Presidente Prudente, SP, Brazil
dc.description.sponsorshipDireccion General de Apoyo al Personal Academico of the Universidad Nacional Autonoma de Mexico, Mexico (DGAPA-UNAM)
dc.description.sponsorshipDGAPA-UNAM
dc.description.sponsorshipDepartments of Statistics of the University of Oxford, UK
dc.description.sponsorshipUniversidade Estadual Paulista Julio de Mesquita Filho - Campus Presidente Prudente, Brazil
dc.description.sponsorshipInstituto de Matematicas of theUniversidad Nacional Autonoma de Mexico, Mexico
dc.description.sponsorshipIdDireccion General de Apoyo al Personal Academico of the Universidad Nacional Autonoma de Mexico, Mexico (DGAPA-UNAM): PAPIIT-IN102713
dc.description.sponsorshipIdDireccion General de Apoyo al Personal Academico of the Universidad Nacional Autonoma de Mexico, Mexico (DGAPA-UNAM): IN102416
dc.format.extent153-184
dc.identifierhttp://dx.doi.org/10.1007/s10651-019-00423-6
dc.identifier.citationEnvironmental And Ecological Statistics. Dordrecht: Springer, v. 26, n. 2, p. 153-184, 2019.
dc.identifier.doi10.1007/s10651-019-00423-6
dc.identifier.issn1352-8505
dc.identifier.urihttp://hdl.handle.net/11449/185824
dc.identifier.wosWOS:000472171700003
dc.language.isoeng
dc.publisherSpringer
dc.relation.ispartofEnvironmental And Ecological Statistics
dc.rights.accessRightsAcesso aberto
dc.sourceWeb of Science
dc.subjectAnisotropic spatial model
dc.subjectBayesian inference
dc.subjectChange-points
dc.subjectMarkov chain Monte Carlo algorithms
dc.subjectNon-homogeneous Poisson process
dc.titleUsing a non-homogeneous Poisson model with spatial anisotropy and change-points to study air pollution dataen
dc.typeArtigo
dcterms.licensehttp://www.springer.com/open+access/authors+rights?SGWID=0-176704-12-683201-0
dcterms.rightsHolderSpringer

Arquivos

Coleções

Artigos - Estatística - FCT