|
Activity Number:
|
2
|
|
Type:
|
Invited
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #302930 |
|
Title:
|
Combining Data for Efficient Prediction of the Spatial Distribution of Iowa Residential Radon Levels
|
|
Author(s):
|
Jun Yan*+ and Mary K. Cowles and Brian J. Smith
|
|
Companies:
|
University of Connecticut and The University of Iowa and The University of Iowa
|
|
Address:
|
215 Glenbrook Road, U-4120, Storrs, CT, 06269,
|
|
Keywords:
|
Bayesian ; geostatistics ; MCMC ; RAMPS
|
|
Abstract:
|
Because of observed associations between radon exposure and the risk of lung cancer and leukemia, radon is an important issue to public-health officials and policy makers. Employing different sampling protocols and collecting different types of data, available data sources presents different advantages and disadvantages in the effort to produce maps of the surface of residential radon concentration. We report here on a pilot study to address these challenges. Using data for the state of Iowa from two sources, we develop realistic Bayesian geostatistical models and feasible computational strategies that enable the production of more precise maps of Iowa residential radon levels than could be produced using either data set alone.
|
