A0841
Title: Estimating population size: The importance of model and estimator choice
Authors: Matthew Schofield - University of Otago (New Zealand) [presenting]
Abstract: The motivation comes from a mark-recapture distance sampling analysis. Unexpectedly large differences were found between Bayesian and frequentist estimates of abundance despite a moderately large number of observations (~600). Further exploration revealed similar sensitivity to estimator choice when focusing on frequentist estimation. To understand these differences, abundance estimation from general mark-recapture models with three estimation strategies (maximum likelihood estimation, conditional maximum likelihood estimation, and Bayesian estimation) is considered for both binomial and Poisson capture-recapture models. It is found that assuming the data have a binomial or multinomial distribution introduces implicit and unnoticed assumptions that are not addressed when fitting with maximum likelihood estimation. This can have an important effect, particularly if the data arise from multiple populations. The results are compared to those of restricted maximum likelihood in linear mixed effects models.