|
Activity Number:
|
410
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #304254 |
|
Title:
|
Multivariate Zero-Inflated Bayesian Spatial Models with Repeated Measurements
|
|
Author(s):
|
Jing Zhang*+ and Chong Z. He
|
|
Companies:
|
Miami University and University of Missouri
|
|
Address:
|
Department of Mathematics and Statistics, Oxford, OH, 45056,
|
|
Keywords:
|
Multivariate ; Bayesian Hierarchical Modeling ; Spatial Analysis ; Zero-inflated ; MCMC ; Repeated Measurements
|
|
Abstract:
|
Zero-inflated data arises in many contexts. In this project, we generalize a Bayesian hierarchical model which deals with the spatial effects, correlation between repeated measurements as well as the excess zeros simultaneously to a multivariate setting. Inference, including the simulation from the posterior distributions and predictions on new locations, is carried out by computationally efficient MCMC techniques. The approach is illustrated via the application to the herbaceous data collected in the Missouri Ozark Forest Ecosystem Project.
|
