A0714
Title: Bayesian modelling and inference for finite populations from process-based superpopulations
Authors: Sudipto Banerjee - UCLA (United States) [presenting]
Abstract: The purpose is to offer some perspectives on Bayesian inference for finite population quantities when the units in the population are assumed to exhibit complex dependencies. More specifically, Bayesian models are developed when the finite population units are assumed to be realizations of a spatial process. With an overview of Bayesian hierarchical models, including some yielding design-based Horvitz-Thompson estimators, dependence is introduced in finite populations, and inferential frameworks are set out for ignorable and nonignorable responses. Multivariate dependencies using graphical models and spatial processes are discussed, and some salient features of two recent analyses of spatially oriented finite populations are presented.
