Hierarchical Bayesian Survival Analysis of Age-Specific Data From Birds' Nests

Abstract

In this thesis, I first present the grassland birds data from Wells(2007) which is used by several different methods of estimating the nest survival rates. The hierarchical Bayesian method from Cao(2009) then is introduced as a new model to estimate nest-specific survival rates with double censored, left-truncated data. I compare two methods and during the comparison, cox-proportional model and intrinsic autoregressive prior are studied In the second half of this thesis, different data analysis methods are introduced, the deviance information criterion is presented and the Bayesian method is compared with the Mayfield method. The hierarchical Bayesian method is relatively new and is a complicated model indeed for those people who are not familiar with the Bayesian and higher dimension of integration. Nevertheless, it is still a valuable statistical tool. The deviance information criterion is a new method of analyses data; users could choose the different priors in order to get different estimating results, therefore it is very applicable in the statistical world.