Abstract
AbstractQuick counts are usually based on a pre-specified, stratified, sampling design. To define the specifications of the design, results from previous elections are used and quantification of the estimation error is obtained. However, in practice, information is gathered gradually with no guarantee of obtaining the whole designed sample. If the original strata are also study domains, incomplete samples worsen the problem because there might be strata with no information at all. To produce partial estimates, we must adequate the estimation procedure. A specific technique based on dynamic clusterings plus a credibility level correction is considered to solve the lack of information in a stratified sampling design, where strata are also study domains. In particular, we use a hierarchical clustering method, based on data from previous elections, to create a poststratification to impute the sufficient statistics for those strata without sample observations available. We present the details of the sampling design and illustrate the results of the federal quick count in the 2021 Mexican election.KeywordsBayesian estimationHierarchical clusteringpoststratificationSufficient statistics
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
