A note on intrinsic conditional autoregressive models for disconnected graphs

Anna Freni-Sterrantino; Massimo Ventrucci; Håvard Rue

doi:10.1016/j.sste.2018.04.002

A note on intrinsic conditional autoregressive models for disconnected graphs

Spat Spatiotemporal Epidemiol. 2018 Aug:26:25-34. doi: 10.1016/j.sste.2018.04.002. Epub 2018 May 23.

Authors

Anna Freni-Sterrantino¹, Massimo Ventrucci², Håvard Rue³

Affiliations

¹ Small Area Health Statistics Unit, Department of Epidemiology and Biostatistics, Imperial College London, United Kingdom. Electronic address: a.freni-sterrantino@imperial.ac.uk.
² Department of Statistics, University of Bologna, Bologna, Italy.
³ CEMSE Division, King Abdullah University of Science and Technology, Saudi Arabia.

PMID: 30390932
DOI: 10.1016/j.sste.2018.04.002

Abstract

In this note we discuss (Gaussian) intrinsic conditional autoregressive (CAR) models for disconnected graphs, with the aim of providing practical guidelines for how these models should be defined, scaled and implemented. We show how these suggestions can be implemented in two examples, on disease mapping.

Keywords: CAR models; Disconnected graph; Disease mapping; Gaussian Markov random fields; INLA; Islands.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Data Interpretation, Statistical
Humans
Italy / epidemiology
Lip Neoplasms / epidemiology
Models, Statistical*
Scotland / epidemiology
Spatio-Temporal Analysis*
Stomach Neoplasms / epidemiology