Online Program Home
My Program

Abstract Details

Activity Number: 147
Type: Invited
Date/Time: Monday, August 1, 2016 : 10:30 AM to 12:20 PM
Sponsor: WNAR
Abstract #318112 View Presentation
Title: Modeling, Estimation, and Interpretion of Covariate Effects in the Presence of Spatial Confounding
Author(s): Ephraim M. Hanks* and Mevin Hooten and Jennifer Hoeting and Erin Schliep
Companies: Penn State University and Colorado State University and Colorado State University and University of Missouri
Keywords: Spatial statistics ; Regularization ; Random Effects

Spatial confounding is caused by collinearity between spatially-smooth fixed effects and spatial random effects. We discuss the problems of identifiability caused by this confounding, and discuss the benefits and pitfalls associated with restricted spatial regression and Bayesian regularization approach to jointly penalize complexity in both fixed and random effects. We also propose an approach to modeling spatial data using mechanistic spatio-temporal generating processes and show how such generating processes can result in spatially confounded data under model misspecification.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association