Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Jan;27(1):250-268.
doi: 10.1177/0962280215627298. Epub 2016 Jul 20.

Spatially-dependent Bayesian model selection for disease mapping

Rachel Carroll  1 , Andrew B Lawson  1 , Christel Faes  2 , Russell S Kirby  3 , Mehreteab Aregay  1 , Kevin Watjou  2
Affiliations

Spatially-dependent Bayesian model selection for disease mapping

Rachel Carroll et al. Stat Methods Med Res. 2018 Jan.

Abstract

In disease mapping where predictor effects are to be modeled, it is often the case that sets of predictors are fixed, and the aim is to choose between fixed model sets. Model selection methods, both Bayesian model selection and Bayesian model averaging, are approaches within the Bayesian paradigm for achieving this aim. In the spatial context, model selection could have a spatial component in the sense that some models may be more appropriate for certain areas of a study region than others. In this work, we examine the use of spatially referenced Bayesian model averaging and Bayesian model selection via a large-scale simulation study accompanied by a small-scale case study. Our results suggest that BMS performs well when a strong regression signature is found.

Keywords: BRugs; Bayesian model averaging; Bayesian model selection; MCMC; R2WinBUGS; spatial.

PubMed Disclaimer

Conflict of interest statement

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Figures

Figure 1
Figure 1
Display of the spatial distribution of simulated covariates per county.
Figure 2
Figure 2
Display of the areas for the partial models.
Figure 3
Figure 3
Models weights associated with the complete models using BMS.
Figure 4
Figure 4
Model probabilities associated with the complete models using BMA.
Figure 5
Figure 5
Model probabilities for the complete model scenarios with the CV term compared to those without the CV term.
Figure 6
Figure 6
Model weights associated with E1PS4F4/S8F8 for BMS.
Figure 7
Figure 7
Model weights associated with E1PS6F6/S9F9 for BMS.
Figure 8
Figure 8
Model probabilities associated with E1PS4F4/S8F8 for BMA.
Figure 9
Figure 9
Model probabilities associated with E1PS6F6/S9F9 for BMA.
Figure 10
Figure 10
Model weights and probabilities associated with the misspecified models fit with E1CS1PF1.
Figure 11
Figure 11
County weights based on the BMS and BMA procedures for the misspecified models.
Figure 12
Figure 12
County weights based on the BMS procedure for the colon cancer example.
Figure 13
Figure 13
County weights based on the BMS procedure for the colon cancer example.
See this image and copyright information in PMC

References

    1. Clyde M, Iversen ES. Bayesian model averaging in the M-open framework. In: Damien P, Dellaportas P, Polson NG, Stephens DA, editors. Bayesian theory and applications. Oxford: Oxford University Press; 2015.
    1. Hoeting JA, Madigan D, Raftery AE, et al. Bayesian model averaging: a tutorial. Stat Sci. 1999;14:382–417.
    1. Lesaffre E, Lawson AB. Bayesian biostatistics. 1. West Sussex, UK: Wiley; 2013. p. 534.
    1. Lunn D, Jackson C, Best N, et al. The BUGS book: a practical introduction to bayesian analysis. 1. Boca Raton, FL: CRC Press; 2013. p. 399.
    1. Thomas A, O’hara B, Ligges U, et al. Making BUGS open. R News. 2006;6:12–17.

Publication types

LinkOut - more resources