Bayesian Capture-Recapture Models that Facilitate Recursive Computing
share
Ecologists increasingly rely on Bayesian capture-recapture models to estimate abundance of wildlife populations. Capture-recapture models account for imperfect detectability in individual-level presence data. A variety of approaches have been used to implement such models, including integrated likelihood, parameter-expanded data augmentation, and combinations of those. Recently proposed conditional specifications have improved the stability of algorithms for fitting capture-recapture models. We arrive at similar conditional specifications of capture-recapture models by considering recursive implementation strategies that facilitate fitting models to large data sets. Our approach enjoys the same computational stability but also allows us to fit the desired model in stages and leverage parallel computing resources. Our model specification includes a component for the capture history of detected individuals and another component for the sample size which is random before observed. We demonstrate this approach using three examples including simulation and two data sets resulting from capture-recapture studies of different species.
READ FULL TEXT
