|
Activity Number:
|
60
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #310413 |
|
Title:
|
spBayes: An R Package for Univariate and Multivariate Hierarchical Point-Referenced Spatial Models
|
|
Author(s):
|
Andrew Finley*+ and Sudipto Banerjee and Brad Carlin
|
|
Companies:
|
The University of Minnesota and The University of Minnesota and The University of Minnesota
|
|
Address:
|
Dept. of Forest Resources, Saint Paul, MN, 55108,
|
|
Keywords:
|
Bayesian inference ; coregionalization ; kriging ; MCMC ; multivariate spatial process
|
|
Abstract:
|
Scientists and investigators in diverse fields often encounter spatially referenced data collected over a fixed set of locations within in a region of study. Such point-referenced (geostatistical) data are often best analyzed with Bayesian hierarchical models. Unfortunately, fitting such models involves computationally intensive Markov chain Monte Carlo methods whose efficiency depends upon the specific problem at hand and often requires extensive coding. Here, we introduce a statistical software package, spBayes, built upon the R statistical computing platform that implements a generalized template encompassing a wide variety of Gaussian spatial process models for univariate as well as multivariate point-referenced data. We discuss the algorithms behind our package and software design considerations. We illustrate with an analysis of forest inventory biomass data.
|
