Individual heterogeneity in studies on marked animals using numerical integration: capture–recapture mixed models

• DOI: 10.1890/09-1903.1
• Publication Date: 2010-04-16T21:01:32Z
• Authors: O. Gimenez; R. Choquet
• Keywords: Mark and recapture; Mixed model; Vital rates

In conservation and evolutionary ecology, quantifying accounting for individual heterogeneity in vital rates of open populations is particular interest. Individual random effects have been used capture–recapture models, adopting a Bayesian framework with Markov chain Monte Carlo (MCMC) to carry out estimation inference. As an alternative, we show how numerical integration via the Gauss‐Hermite quadrature (GHQ) can be efficiently approximate model likelihood effects. We compare performance two approaches (MCMC vs. GHQ) finite mixture models using examples, including data on European Dippers Sociable Weavers. Besides relying standard statistical tools, GHQ was found faster than MCMC simulations. Our approach implemented program E‐SURGE. Overall, mixed (CR2Ms), either approximation or simulations, potential important applications population biology.