Capture-Recapture Estimation for Elusive Events with Two Lists

Abstract

The application of capture-recapture methods to the estimation of population parameter for epidemiologic and demographic event has been growing recently. This paper presents a robust capture-recapture model for estimating the size of elusive epidemiologic event. The proposed estimator , the Petersen estimator  and another estimator  were compared using the Akaike Information Criterion (AIC) and the Mean Absolute Deviation (MAD) through simulation studies. Both the AIC and MAD revealed that  is a better and robust estimator. It was discovered that  under estimates the total elusive population N,  over estimate N while  was always consistent and performs better than the other two.  is particularly better with lower recaptures (‘relisting’)  which is the case with elusive events. The Petersen  breaks down in the presence of elusiveness. The other estimator,  always over estimates, except in few cases. It is therefore recommended that the proposed estimator  be used for estimating dual system elusive events.

Keywords: Trap Response; Dual System Estimation; Elusive events; Petersen estimator; capture-recapture.