|
Activity Number:
|
349
- Longitudinal, Spatial, and Bayesian Methods
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #330002
|
Presentation
|
|
Title:
|
Time-Stratified LOESS Smoothers for Estimating and Testing Temporal Heterogeneity in Spatial Risk Patterns
|
|
Author(s):
|
Yannan Tang*
|
|
Companies:
|
UCI
|
|
Keywords:
|
generalized additive models;
spatiotemporal effects;
permutation test;
LOESS;
spatial epidemiology
|
|
Abstract:
|
Public health researchers often aim to quantify spatial heterogeneity in disease occurrence in order to identify potential health disparities. In these studies, spatial information is becoming increasingly available at the person-level. Generalized additive models (GAMs) provide a framework for smoothing risk estimates over space while adjusting for confounding factors. Longitudinal data tracking disease indicators over time are becoming more prevalent. As such, modeling variation in spatial effects over time is of increased interest. We propose a GAM framework that incorporates temporally stratified LOESS smoothers to estimate geospatial effects at differing time points while borrowing information over time on confounding effects. We further develop a permutation testing strategy to test for temporal heterogeneity in spatial effects using the proposed time-stratified estimates. Monte Carlo studies are presented to demonstrate the advantage of the method when time-varying spatial patterns exist. Our empirical results further suggest that the proposed testing procedure yields high power for detecting temporal heterogeneity in spatial effects while controlling type I error.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2018 program
|
