Sélection de la langue

Recherche


Best feasible unbiased prediction for multi-source data system with an application for multiplicative benchmarking / Zhao-Guo Chen and Estela Bee Dagum. : CS11-617/98-9E-PDF

"Information about a socio-economic variable of interest (usually, one or a group of time series) often originates from several sources none of which is complete and/or accurate. A stepwise approach is developed here for predicting the variable of interest by using the data from source to source focusing on minimizing the variances of prediction errors. This paper also reviews some BLUP (the best linear unbiased prediction) theory and shows that the stepwise approach proposed here can give better predictions than BLUP for nonlinear models. As an important application, a nonlinear benchmarking formula for a multiplicative model is derived"--Summary.

Lien permanent pour cette publication :
publications.gc.ca/pub?id=9.840289&sl=1

Renseignements sur la publication
Ministère/Organisme
  • Canada. Statistics Canada. Methodology Branch.
TitreBest feasible unbiased prediction for multi-source data system with an application for multiplicative benchmarking / Zhao-Guo Chen and Estela Bee Dagum.
Titre de la série
  • Working paper ; 98-9
Type de publicationMonographie - Voir l'enregistrement principal
Langue[Anglais]
FormatTexte numérique
Document électronique
Note(s)
  • Digitized edition from print [produced by Statistics Canada].
  • "Working paper No. BSMD-98-009E."
  • "October 1998."
  • Includes bibliographic references.
  • Summary also in French.
Information sur la publication
  • [Ottawa] : Statistics Canada, 1998.
Auteur / Contributeur
  • Chen, Zhao-Guo,1943-
  • Dagum, Estela Bee.
Description23 p.
Numéro de catalogue
  • CS11-617/98-9E-PDF
Numéro de catalogue du ministère11-617 no. 98-09
Descripteurs
Demander des formats alternatifs
Pour demander une publication dans un format alternatif, remplissez le formulaire électronique des publications du gouvernement du Canada. Utilisez le champ du formulaire «question ou commentaire» pour spécifier la publication demandée.

Détails de la page

Date de modification :