Bayesian model assisted design-based estimators of the size, total and mean of a hard-to-reach population from a link-tracing sample with initial cluster sample

Articles and reports: 11-522-X202200100015
Description: We present design-based Horvitz-Thompson and multiplicity estimators of the population size, as well as of the total and mean of a response variable associated with the elements of a hidden population to be used with the link-tracing sampling variant proposed by Félix-Medina and Thompson (2004). Since the computation of the estimators requires to know the inclusion probabilities of the sampled people, but they are unknown, we propose a Bayesian model which allows us to estimate them, and consequently to compute the estimators of the population parameters. The results of a small numeric study indicate that the performance of the proposed estimators is acceptable.
Issue Number: 2022001
Author(s): Félix-Medina, Martin
Main Product: Statistics Canada International Symposium Series: Proceedings
Format Release date More information
PDF March 25, 2024

Related information

Subjects and keywords

Subjects

Keywords

Date modified: