Please use this identifier to cite or link to this item: https://doi.org/10.1093/biomet/asn074
DC FieldValue
dc.titleModel checking in regression via dimension reduction
dc.contributor.authorXia, Y.
dc.date.accessioned2014-10-28T05:13:09Z
dc.date.available2014-10-28T05:13:09Z
dc.date.issued2009-03
dc.identifier.citationXia, Y. (2009-03). Model checking in regression via dimension reduction. Biometrika 96 (1) : 133-148. ScholarBank@NUS Repository. https://doi.org/10.1093/biomet/asn074
dc.identifier.issn00063444
dc.identifier.urihttp://scholarbank.nus.edu.sg/handle/10635/105222
dc.description.abstractLack-of-fit checking for parametric and semiparametric models is essential in reducing misspecification. The efficiency of most existing model-checking methods drops rapidly as the dimension of the covariates increases. We propose to check a model by projecting the fitted residuals along a direction that adapts to the systematic departure of the residuals from the desired pattern. Consistency of the method is proved for parametric and semiparametric regression models. A bootstrap implementation is also discussed. Simulation comparisons with several existing methods are made, suggesting that the proposed methods are more efficient than the existing methods when the dimension increases. Air pollution data from Chicago are used to illustrate the procedure. © 2009 Biometrika Trust.
dc.sourceScopus
dc.subjectBootstrap
dc.subjectCrossvalidation
dc.subjectGoodness-of-fit
dc.subjectKernel smoothing
dc.subjectSemiparametric model
dc.subjectSingle-index model
dc.typeArticle
dc.contributor.departmentSTATISTICS & APPLIED PROBABILITY
dc.description.doi10.1093/biomet/asn074
dc.description.sourcetitleBiometrika
dc.description.volume96
dc.description.issue1
dc.description.page133-148
dc.description.codenBIOKA
dc.identifier.isiut000263406100010
Appears in Collections:Staff Publications

Show simple item record
Files in This Item:
There are no files associated with this item.

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.