This paper presents a Bayesian analysis of mark-recapture-recovery data on Soay sheep. A reversible-jump Markov chain Monte Carlo technique is used to determine age-classes of common survival, and to model the survival probabilities in those age classes using logistic regression. This involves both environmental and individual covariates, as well as random effects. Auxiliary variables are used to impute missing covariates measured on individual sheep. The Bayesian approach suggests different models from those previously obtained for these data using classical statistical methods. Following model-averaging, features are identified which were not previously detected, and which are of ecological importance.
age-classes; annual survival; auxiliary variables; Bayesian p-values; goodness-of-fit; logistic regression; mark-recapture-recovery; mixed models; model-averaging; North Atlantic Oscillation; random effects; reversible jump Markov chain Monte Carlo; senescence; Soay sheep; trans-dimensional simulated annealing
Appeared as King, R., Brooks, S. P., Morgan, B. J. T. and Coulson, T. (2006) "Factors Influencing Soay Sheep Survival: A Bayesian Analysis". Biometrics 62 pp211-220