Bayesian Modeling of Age-Specific Survival in Nesting Studies Under Dirichlet Priors |
| |
Authors: | Chong Z He Dongchu Sun Yolande Tra |
| |
Institution: | Department of Statistics, University of Missouri-Columbia, 65211, USA. chong@stat.missouri.edu |
| |
Abstract: | There has been much work done in nest survival analysis using the maximum likelihood (ML) method. The ML method suffers from the instability of numerical calculations when models having a large number of unknown parameters are used. A Bayesian approach of model fitting is developed to estimate age-specific survival rates for nesting studies using a large class of prior distributions. The computation is done by Gibbs sampling. Some latent variables are introduced to simplify the full conditional distributions. The method is illustrated using both a real and a simulated data set. Results indicate that Bayesian analysis provides stable and accurate estimates of nest survival rates. |
| |
Keywords: | Age-specific survival Bayesian estimation Censoring Gibbs sampling Latent variable Nest study Truncation |
|