作者: John N. Darroch , Stephen E. Fienberg , Gary F. V. Glonek , Brian W. Junker
DOI: 10.1080/01621459.1993.10476387
关键词:
摘要: Abstract A central assumption in the standard capture-recapture approach to estimation of size a closed population is homogeneity “capture” probabilities. In this article we develop an that allows for varying susceptibility capture through individual parameters using variant Rasch model from psychological measurement situations. Our requires additional recapture. context census undercount estimation, requirement amounts use second independent sample or alternative data source be matched with and Post-Enumeration Survey (PES) data. The models provide mechanism separating out dependence between PES induced by heterogeneity. resulting take form incomplete 23 contingency table, describe how estimate expected values observable cells table log-linear quasi-symmetry models. projection these estimates onto unobserved...