School Seminars and Colloquia

Modelling Capture-recapture Experiments in Closed Population

Statistics Seminar

by Ning Li


Institution: The University of Melbourne
Date: Tue 8th April 2008
Time: 1:15 PM
Location: Room 213, Richard Berry Bldg, The University of Melbourne

Abstract: We consider the problem of modelling capture-recapture experiments to estimate the size of a closed animal population, where the capturebility may vary among individuals. We examine the modelling aspects of some
commonly used statistical approaches in the literature, their associated
data types and the possibilities for extension to accommodate other data
types. We argue that not only the observed data but also the underlying process that generated the data are relevant in capture-recapture
modelling. And both should be explicitly formulated in a model in order to estimate the population size appropriately. We derive a general approach to constructing such capture-recapture models within the framework of missing-data formulation. We discuss some frequently encountered models as special cases of this general formulation; and we explore the scope that
the missing-data formulation has in estimating animal population size when the capture-ability is individual-dependent. Throughout, we emphasize the
role that basic probability rules have in modelling capture-recapture data.

For More Information: Dr .Guoqi Qian g.qian@ms.unimelb.edu.au