A non-linear hierarchical modelling approach for census undercoverage estimation: CS92-0049/2000E-PDF

"Area-level nonlinear mixed effects models are considered in this paper for Canada census undercoverage estimation. We fit an area-level nonlinear mixed effects model to the province-level undercoverage survey estimates. In particular, the sampling model is based on the survey estimate of the undercoverage count, and the linking model is a log-linear model for the undercoverage rate. A full hierarchical Bayes (HB) approach is developed to obtain the posterior estimates of the census undercoverage using Markov Chain Monte Carlo (MCMC) sampling methods. Our result shows that the proposed method can provide efficient model-based estimates. Analysis of model fitting is also presented using posterior predictive distributions, and the corresponding result indicates that the proposed model fits the data quite well."--Abstract.

Permanent link to this Catalogue Record: What is a permanent link?
http://publications.gc.ca/pub?id=9.839753&sl=0
MARC XML Format   MARC HTML Format

Department/Agency Statistics Canada. Methodology Branch. Household Survey Methods Division.
Title A non-linear hierarchical modelling approach for census undercoverage estimation
Publication Type Monograph
Language [English]
Format Electronic
Electronic Document

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact the authoring department to request a format other than those available.

We invite you to consult the Frequently Asked Questions page for additional information regarding the Archived Content notice.


Having trouble opening this document?

Note: The URLs contained in this/these document(s) may no longer be functional
Note Digitized edition from print [produced by Statistics Canada]. Title from caption. Copy of an article from Proceedings of the Survey Methods Section, 2000. Includes abstract in French.
Date 2000]
Number of Pages p. 185-190
Catalogue Number
  • CS92-0049/2000E-PDF
Subject Terms Census, Methodology, Statistical analysis