A shared neighbor conditional autoregressive model for small area spatial data

Abstract

The use of conditional autoregressive (CAR) models for spatial effects is commonplace, especially when dealing with aggregated count data in health studies. CAR models are convenient and relatively easy to implement but suffer from the fact that they have limited flexibility in modeling correlation. We introduce a new CAR model that can accommodate different neighborhood features (including shared neighbors). Further, we examine via simulation how this model performs in comparison with standard CAR models. We also consider the application to a small area health data example. Copyright © 2015 John Wiley & Sons, Ltd.

Modelling and predicting the spatio-temporal spread of COVID-19, associated deaths and impact of key risk factors in England

“…A previously developed Bayesian space–time SEIR formulation 8 was then applied to assess the spatiotemporal variability of COVID-19 transmission at small area scale (MSOA) and by week in England, by accounting for the modelled transmission dynamics of the pathogen, inherent spatial–temporal correlation in the data, and important contextual risk factors for both COVID-19 cases and deaths. We also assessed the sum of cases in shared neighbours in the preceding time point as an additional parameter in our model to further assess the dependence of caseloads in adjoined areas 27 . Based on the available data provided we the starting study week was denoted as week 9 of 2020 (starting 24 February 2020), with available case data at the time of extraction for this analysis available up to week 34 (starting 17 August 2020) and death data available up to week 26 (starting 22June 2020).…”