Header

UZH-Logo

Maintenance Infos

A primer on disease mapping and ecological regression using INLA


Schrödle, B; Held, L (2011). A primer on disease mapping and ecological regression using INLA. Computational Statistics, 26(2):241-258.

Abstract

Within this paper spatial and spatio-temporal disease mapping models are reviewed and applied to reported Coxiellosis cases among cows in Switzerland, 2005-2008. Furthermore, an ecological regression is conducted using a linear and nonparametric association between the number of stillborn calves and the reported Coxiellosis cases within one Swiss region. As a tool for Bayesian inference integrated nested Laplace approximations (INLA) are used. INLA is a promising alternative to Markov chain Monte Carlo (MCMC) methods which is supposed to provide very accurate results within much less computational time. From a user's point of view INLA can easily be applied using an R package called INLA. Hence, it is shown how spatial and spatio-temporal models can be specied and run using INLA; some of the respective R code is also shown. Additionally, computational aspects ascomputational time, accuracy of the results and usability of the R packageare addressed. As a result it was found that the INLA package is easy to handle for this class of models although INLA is a very complex numerical algorithm. Furthermore, useful tools for model choice can be obtained as a standard output.

Abstract

Within this paper spatial and spatio-temporal disease mapping models are reviewed and applied to reported Coxiellosis cases among cows in Switzerland, 2005-2008. Furthermore, an ecological regression is conducted using a linear and nonparametric association between the number of stillborn calves and the reported Coxiellosis cases within one Swiss region. As a tool for Bayesian inference integrated nested Laplace approximations (INLA) are used. INLA is a promising alternative to Markov chain Monte Carlo (MCMC) methods which is supposed to provide very accurate results within much less computational time. From a user's point of view INLA can easily be applied using an R package called INLA. Hence, it is shown how spatial and spatio-temporal models can be specied and run using INLA; some of the respective R code is also shown. Additionally, computational aspects ascomputational time, accuracy of the results and usability of the R packageare addressed. As a result it was found that the INLA package is easy to handle for this class of models although INLA is a very complex numerical algorithm. Furthermore, useful tools for model choice can be obtained as a standard output.

Statistics

Citations

Dimensions.ai Metrics
47 citations in Web of Science®
54 citations in Scopus®
87 citations in Microsoft Academic
Google Scholar™

Altmetrics

Downloads

9 downloads since deposited on 13 Dec 2010
7 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:04 Faculty of Medicine > Epidemiology, Biostatistics and Prevention Institute (EBPI)
Dewey Decimal Classification:610 Medicine & health
Language:English
Date:2011
Deposited On:13 Dec 2010 15:28
Last Modified:21 Sep 2018 15:15
Publisher:Springer
ISSN:0943-4062
OA Status:Green
Publisher DOI:https://doi.org/10.1007/s00180-010-0208-2

Download

Download PDF  'A primer on disease mapping and ecological regression using INLA'.
Preview
Content: Published Version
Language: English
Filetype: PDF (Nationallizenz 142-005)
Size: 480kB