A Bayesian latent process spatiotemporal regression model for areal count data
A Bayesian latent process spatiotemporal regression model for areal count data
Model-based approaches for the analysis of areal count data are commonplace in spatiotemporal analysis. In Bayesian hierarchical models, a latent process is incorporated in the mean function to account for dependence in space and time. Typically, the latent process is modelled using a conditional autoregressive (CAR) prior. The aim of this paper is to offer an alternative approach to CAR-based priors for modelling the latent process. The proposed approach is based on a spatiotemporal generalization of a latent process Poisson regression model developed in a time series setting. Spatiotemporal dependence in the autoregressive model for the latent process is modelled through its transition matrix, with a structured covariance matrix specified for its error term. The proposed model and its parameterizations are fitted in a Bayesian framework implemented via MCMC techniques. Our findings based on real-life examples show that the proposed approach is at least as effective as CAR-based models.
Autoregressive latent process, Bayesian inference, Conditional autoregressive prior, Markov Chain Monte Carlo, Spatiotemporal areal count data
25-37
Utazi, Chigozie Edson
97af8901-3d52-46c5-8d16-e68e29057aa6
Afuecheta, Emmanuel
1d801ab5-1e19-42d8-84b7-6d6a5cc5c565
Nnanatu, Chibuzor Christopher
24be7c1b-a677-4086-91b4-a9d9b1efa5a3
1 June 2018
Utazi, Chigozie Edson
97af8901-3d52-46c5-8d16-e68e29057aa6
Afuecheta, Emmanuel
1d801ab5-1e19-42d8-84b7-6d6a5cc5c565
Nnanatu, Chibuzor Christopher
24be7c1b-a677-4086-91b4-a9d9b1efa5a3
Utazi, Chigozie Edson, Afuecheta, Emmanuel and Nnanatu, Chibuzor Christopher
(2018)
A Bayesian latent process spatiotemporal regression model for areal count data.
Spatial and Spatio-temporal Epidemiology, 25 (6), .
(doi:10.1016/j.sste.2018.01.003).
Abstract
Model-based approaches for the analysis of areal count data are commonplace in spatiotemporal analysis. In Bayesian hierarchical models, a latent process is incorporated in the mean function to account for dependence in space and time. Typically, the latent process is modelled using a conditional autoregressive (CAR) prior. The aim of this paper is to offer an alternative approach to CAR-based priors for modelling the latent process. The proposed approach is based on a spatiotemporal generalization of a latent process Poisson regression model developed in a time series setting. Spatiotemporal dependence in the autoregressive model for the latent process is modelled through its transition matrix, with a structured covariance matrix specified for its error term. The proposed model and its parameterizations are fitted in a Bayesian framework implemented via MCMC techniques. Our findings based on real-life examples show that the proposed approach is at least as effective as CAR-based models.
Text
Accepted manuscript
- Accepted Manuscript
More information
Accepted/In Press date: 23 January 2018
e-pub ahead of print date: 2 February 2018
Published date: 1 June 2018
Additional Information:
Erratum regarding missing Declaration of Competing Interest statements in previously published articles Spatial and Spatio-temporal Epidemiology, Volume 36, February 2021, Pages 100400
Keywords:
Autoregressive latent process, Bayesian inference, Conditional autoregressive prior, Markov Chain Monte Carlo, Spatiotemporal areal count data
Identifiers
Local EPrints ID: 417929
URI: http://eprints.soton.ac.uk/id/eprint/417929
ISSN: 1877-5845
PURE UUID: f17d60d4-e168-45f6-9be6-8524f35ff25e
Catalogue record
Date deposited: 16 Feb 2018 17:30
Last modified: 13 Jun 2024 04:03
Export record
Altmetrics
Contributors
Author:
Chigozie Edson Utazi
Author:
Emmanuel Afuecheta
Author:
Chibuzor Christopher Nnanatu
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics