Activity Number:
|
422
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistics and the Environment
|
Abstract - #305778 |
Title:
|
Hierarchical Spatial Modeling of Air Pollution Exposure and Measurement Error
|
Author(s):
|
Simone Gray*+ and Alan E. Gelfand and Marie Lynn Miranda and Sharon Edwards
|
Companies:
|
Duke University and Duke University and Duke University and Duke University
|
Address:
|
Box 90251 , Durham, NC, 27708,
|
Keywords:
|
measurement error ; air pollution ; exposure assessment ; birth weight
|
Abstract:
|
Air pollution (AP) measurements from the closest monitor are commonly used as a proxy for personal exposure. This introduces measurement error (ME) into a model. This talk attempts to better understand the relationship between exposure to AP and pregnancy outcomes, while accounting for the associated ME. Induced by a process model specification for exposure reflecting sparsity of monitoring sites, we construct a spatial model that allows uncertainty in exposure to increase as the distance between maternal residence and the location of the closest monitor increases. We illustrate with air quality data from the EPA. We account for missing daily data by assuming that error increases as the time from the nearest recorded measurement increases and incorporate that error term into a temporal component of the model. The statistical analyses are implemented using Bayesian hierarchical modeling.
|