EcoSta 2019: Start Registration
View Submission - EcoSta2019
Title: Computational approaches to spatio-temporal surveillance of small area health data Authors:  Andrew Lawson - Medical University of South Carolina (United States) [presenting]
Abstract: Some approaches to ST surveillance when MC sampling is used for posterior characterization are reviewed. With infectious disease as the main focus, we will consider the use of a variety of metrics, which are either residual based, or based on posterior functionals, such as SCPO, SKL, and surveillance residuals. We propose using new combinations of these metrics which include partial prediction using two stage models. We also explore the use of directional resultants in attempting to predict the spatial pathways for future infection in an infectious disease context. If time permits, we will also contrast the use of sequential MC as compared to conventional McMC when metrics are used.