Dual and multiple system estimation: fully observed and incomplete covariates

Peter G.M. van der Heijden, Maarten Cruyff, Joe Whittaker, B.F.M. Bakker, Paul A. Smith

Research output: Chapter in Book / Report / Conference proceedingChapterAcademic

Abstract

This chapter presents the (in)variance of population size estimates derived from loglinear models that include covariates. Including covariates in log-linear models of population registers improves population size estimates for two reasons. Firstly, it is possible to take heterogeneity of inclusion probabilities over the levels of a covariate into account; and secondly, it allows subdivision of the estimated population by the levels of the covariates, giving insight into characteristics of individuals that are not included in any of the registers. The issue of whether/not marginalizing the full table of registers by covariates over one or more covariates leaves the estimated population size estimate invariant, is intimately related to collapsibility of contingency tables. With information from two registers it is shown that population size invariance is equivalent to the simultaneous collapsibility of each margin consisting of one register and the covariates. Covariates that are collapsible are called passive, to distinguish them from covariates that are not collapsible and are termed active.
Original languageEnglish
Title of host publicationCapture-Recapture Methods for the Social and Medical Sciences
EditorsDankmar Böhning, John Bunge, Peter van der Heijden
PublisherCRC Press, Boca Rata
Chapter15
Pages215-230
Number of pages16
ISBN (Electronic)9781315151939
ISBN (Print)9781498745314, 9781032096698
DOIs
Publication statusPublished - 2017

Fingerprint

Dive into the research topics of 'Dual and multiple system estimation: fully observed and incomplete covariates'. Together they form a unique fingerprint.

Cite this