Estimating the Population Parameters for Open Population Capture-Recapture Data Using Generalized Estimating Equations Approach

Abstract

In this study, a Generalized Estimating Equation (GEE) approach accounting for heterogeneity, behavioural effects and environmental effects to estimate the probability of capture, the survival probability and the population size for open population capture-recapture experiment is proposed. A non-spatial open population capture-recapture data was simulated using two different trapping occasions and the number of monte-carlo simulation .The GEE estimates for behavioural/heterogeneity covariates, behavioural/heterogeneity/environmental covariates model under different working correlation. The study revealed that, the 4 recapture occasions yields less QIC, which implies that, the less the capture occasion the better the estimates. The results also show that, the estimated population size decreases when the number of capture occasion is increased. Comparison between heterogeneity/behavioural model and heterogeneity/behavioural/environmental model for the simulations and real life data shows that, the heterogeneity/behavioural/environmental model performs better in estimating the probability of capture the survival probability and the population size, because of the lower standard error. The study concludes that, environmental covariates be considered when estimating the probability of capture, the survival probability and the estimated population size.