Two-Stage Methods for Regression with Spatially Misaligned Data

• Activity Number: 21
• Type: Topic Contributed
• Date/Time: Sunday, July 29, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Kenneth K Lopiano*+ and Linda J Young and Carol Gotway
• Companies: University of Florida and University of Florida and CDC
• Keywords: spatial misalignment ; berkson error ; psuedolikelihood ; linear models ; generalized linear models

Abstract

Researchers in the fields of climate change, environmental risk assessment, and public health often augment their data collection with existing data or work entirely with existing data from multiple sources. When the datasets have a spatial component, the datasets are often spatially misaligned. When the datasets are spatially misaligned, the data must be combined using a common set of geographical units before relationships can be assessed. If the datasets are combined using kriging, then the problem can be recast in a measurement error framework. In this work, we present two-stage likelihood based methods for estimating parameters in both linear and generalized linear regression models. The statistical properties of the estimators are presented and simulation studies are used to illustrate the performance of the methodology.