Structurally missing data problems in multiple list capture–recapture data

Abstract

Multiple-list capture–recapture data can be used to estimate the size of a population. In this manuscript two problems are studied and solved using a common solution. The first problem is that the lists refer to different but overlapping populations. An example is that lists refer to different but overlapping regions, different but overlapping periods in time, or different but overlapping age groups. The second problem is that each list has a set of covariates and the sets of covariates are not identical. By considering both problems as missing data problems, a solution is obtained through the EM algorithm. This approach is illustrated by two examples.