Model-based estimation of the link between the daily survival probability and a time-varying covariate, application to mosquitofish survival data

摘要: The survival probability in a group of individuals may evolve time due to the influence time-varying covariate. In this paper we present model-based approach allowing estimation functional link between and covariate when data are grouped time-period censored. is based on an underlying model consisting non-stationary Markov processes describing individuals. aggregated at level handle structure censoring. aggregation yields generalized non-linear mixed model. Then, Bayesian procedure allows parameters description This applied order explore relationship daily mosquitofish (Gambusia holbrooki) their lengths (small die with higher rate than large ones because they more affected by predation, cannibalism environmental stress).