Application of Administrative Record-based Triple-System Estimator for Estimating Net Errors in Population Census Data
LIU Yuhuan
HUANG Yanhua
HU Guihua
WEN Ting
Abstract：This paper proposes a triple-system estimator as a superior alternative estimation method to the widely used dual-system estimator to address the problem of interaction bias and biases in the estimation of the overall actual population and the net census error.The triple-system estimator is constructed using the census list,the post-enumeration survey list,and the administrative record list.Literature interpretation,field research,sample estimation,and mathematical modeling are used to examine the triple-system estimator's suitability for census net error estimation.The results show that the triple-system estimator relaxes the restriction of the statistical relationship of the three lists and improves the accuracy of the net error estimation by using two kinds of auxiliary information simultaneously,and thus has a higher estimation accuracy.The innovation lies in the use of micro data obtained from a self-survey,a comprehensive demonstration of the calculation process involved,and a data comparison of the precision of the triple-system estimator,the dual-system estimator,and the single-system estimator.
Keywords：census quality evaluationtriple-system estimatorpost-stratificationstratified jack-knife method
Publication Date：2024-03-30
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：14( 85-98 )