Martin Ridout
Work pages
Estimating closed population size from capture-recapture data
This page provides R functions that can be downloaded to fit some models of capture heterogeneity to estimate N, the size of a closed population, from capture-recapture data. This is joint work with Byron Morgan.
In the most general model, captures are modelled as a mixture of binomial and beta-binomial counts. This generalises the beta-binomial model considered, for example, by Dorazio and Royle [Ref 1] and the (two-component form of) finite mixture model considered, for example, by Pledger [Ref 2]. Our new model is described in Refs [3] and [4] below, which also provide other references to the existing literature.
There are important issues of identifiability in models of capture heterogeneity, see for example [Ref 5]. Our model does not, of course, resolve these problems, but it does offer a broad and flexible class of models within which to investigate and compare estimates of N.
A user guide provides information about the software - how to use it and some details about computational methods. To use the software, you will need to download the file estimateN.r into a suitable directory on your computer. This contains all the functions needed to fit the models. Ordinarily, you should not need to look at the R code in this file unless you want to see the details of what's going on. You'll need a reasonably up-to-date version of R to get this to work (it certainly works with 2.6.1 onwards).
Here are a couple of example files that illustrate the use of the software:
All the data sets used in Refs [3] and [4] are in this file.
References
[1] Dorazio, R.M. and Royle, J.A. (2003). Mixture models for estimating the size of a closed population when capture rates vary among individuals. Biometrics, 59, 351-364.
[2] Pledger, S. (2000). Unified maximum likelihood estimates for closed capture-recapture models using mixtures. Biometrics, 56, 434-442.
[3] Morgan, B.J.T. and Ridout, M.S. (2007). A new mixture model for capture heterogeneity. Applied Statistics, to appear.
[4] Morgan, B.J.T. and Ridout, M.S. (2007). Estimating N: a robust approach to capture heterogeneity. under revision
[5] Link, W.A. (2003). Nonindentifiability of population size from capture-recapture data with heterogeneous detection probabilities. Biometrics, 59, 1123-1130.
This page last modified 4 June 2008.
