|
Abstract:
|
Model-based small area estimation is frequently used in conjunction with survey data in order to establish estimates for under-sampled or unsampled geographies. These models can be specified at either the area-level, or the unit-level, but unit-level models often offer potential advantages such as more precise estimates and easy spatial aggregation. In modeling small areas at the unit level, challenges often arise as a consequence of the informative sampling used to collect the survey data. Similar to area-level models, latent Gaussian process (LGP) models can be used within a Bayesian framework to take advantage of underlying dependencies. Nevertheless, LGP models often present computational difficulties in high-dimensional settings. We explore the use of both latent Gaussian processes as well as new distribution theory to model multivariate non-Gaussian responses under informative sampling. We compare the utility and computational feasibility of both models via a simulation study and an application to American Community Survey (ACS) data.
|
