|
Activity Number:
|
491
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 10, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
ENAR
|
| Abstract - #305155 |
|
Title:
|
Bayesian Nonparametric Spatial and Spatio-Temporal Models for Disease Incidence Data
|
|
Author(s):
|
Athanasios Kottas*+ and Jason Duan and Alan E. Gelfand
|
|
Companies:
|
University of California, Santa Cruz and Duke University and Duke University
|
|
Address:
|
School of Engineering, MS: SOE2, Santa Cruz, CA, 95064,
|
|
Keywords:
|
areal unit spatial data ; Dirichlet process mixture models ; disease mapping ; dynamic spatial process model ; Gaussian process ; spatial Dirichlet process
|
|
Abstract:
|
We propose Bayesian nonparametric approaches to modeling disease incidence data, which are typically available as rates or counts for specified regions and collected over time. Within a hierarchical framework, we use a spatial Dirichlet process as the model for the surface of spatial random effects. We then block average the spatial Dirichlet process to the areal units determined by the regions in the study to obtain a prior model for the finite dimensional distribution of the spatial random effects. Moreover, we employ a dynamic formulation for the spatial random effects to extend the model to spatio-temporal settings. We illustrate the methodology with simulated data and a dataset on lung cancer incidences for all 88 counties in Ohio over an observation period of 21 years.
|
