Model selection in Cormack-Jolly-Seber mark-recapture investigations

Release Date:

Return to ASC Quantitative Ecology

Cormack-Jolly-Seber (CJS) mark-recapture models are widely utilized in ecology, providing estimates of vital rates and abundance that are difficult to obtain using other means. CJS models are constructed from combinations of sub-models for survival and recapture probabilities, and several explanatory covariates for each set of probabilities may be under consideration in some investigations. The number of candidate models may exceed feasible limits and investigators may wish to implement a method of reducing the size of the model set. Stepwise methods of model selection that are widely utilized (e.g., Lebreton et al. 1992) can confound information of survival and recapture probabilities (Catchpole at al. 2004) and produce biased assessments of the importance of explanatory covariates (Doherty et al. 2012). Bromaghin et al. (2013) present a method of reducing the number of candidate models that is objective, considers survival and recapture probabilities independently, and provides an unbiased assessment of covariate importance. Their method has been incorporated into the R mark-recapture package mra: http://cran.r-project.org/web/packages/mra/index.html

Related Publications

Bromaghin, J. F., T. L. McDonald, and S. C. Amstrup. 2013. Plausible combinations: An improved method to evaluate the covariate structure of Cormack-Jolly-Seber mark-recapture models. Open Journal of Ecology 3:11-22. doi:10.4236/oje.2013.31002

Catchpole, E.A., Fan, Y., Morgan, B.J.T., Clutton-Brock, T. and Coulson, T. 2004. Sexual dimorphism, survival and dispersal in red deer. Journal of Agricultural, Biological, and Environmental Statistics 9:1-26. doi:10.1198/1085711043172

Doherty, P., White, G. and Burnham, K. 2012. Comparison of model building and selection strategies. Journal of Ornithology 152:317-323. doi:10.1007/s10336-010-0598-5

Lebreton, J.D., Burnham, K.P., Clobert, J. and Anderson, D.R. 1992. Modeling survival and testing biological hypotheses using marked animals—A unified approach with case-studies. Ecological Monographs 62:67-118. doi:10.2307/2937171