Bayesian disease mapping, yet if undeniably useful to describe variation in risk over time and space, comes with the hurdle of prior elicitation on hard-to-interpret random effect precision parameters. We introduce a reparametrized version of the popular spatio-temporal interaction models, based on Kronecker product intrinsic Gaussian Markov random fields, that we name the variance partitioning model. The variance partitioning model includes a mixing parameter that balances the contribution of the main and interaction effects to the total (generalized) variance and enhances interpretability. The use of a penalized complexity prior on the mixing parameter aids in coding prior information in an intuitive way. We illustrate the advantages of the variance partitioning model using two case studies.
DigglePJChetwyndAGHggkvistR, et al. Second-order analysis of space–time clustering. Statist Methods Med Res1995; 4: 124–136. PMID: 7582201.
2.
Martinez-BeneitoMABotella-RocamoraP. Disease mapping: From foundations to multidimensional modeling. 1st ed. Boca Raton, FL: Chapman and Hall/CRC, 2019.
3.
LawsonAB. Bayesian disease mapping: hierarchical modeling in spatial epidemiology. 3rd ed. Boca Raton, FL: Chapman and Hall/CRC, 2018.
4.
MacNabYC. Bayesian disease mapping: past, present, and future. Spat Stat 2022: 100593. DOI:10.1016/j.spasta.2022.100593.
5.
WakefieldJ. Disease mapping and spatial regression with count data. Biostatistics2007; 8: 158–183.
6.
Knorr-HeldLBesagJ. Modelling risk from a disease in time and space. Stat Med1998; 17: 2045–2060.
7.
AbellanJJRichardsonSBestN. Use of space–time models to investigate the stability of patterns of disease. Environ Health Perspect2008; 116: 1111–1119.
8.
Knorr-HeldL. Bayesian modelling of inseparable space–time variation in disease risk. Stat Med2000; 19: 2555–2567.
9.
WallerLACarlinBPXiaH, et al. Hierarchical spatio-temporal mapping of disease rates. J Am Stat Assoc1997; 92: 607–617.
10.
BernardinelliLClaytonDPascuttoC, et al. Bayesian analysis of space–time variation in disease risk. Stat Med1995; 14: 2433–2443.
11.
AldstadtJ. An incremental Knox test for the determination of the serial interval between successive cases of an infectious disease. Stoch Environ Res Risk Assess2007; 21: 487–500.
12.
RobertsonCNelsonTAMacNabYC, et al. Review of methods for space–time disease surveillance. Spatial Spatio-temporal Epidemiol2010; 1: 105–116.
13.
LindgrenFRueH. On the second-order random walk model for irregular locations. Scand J Stat2008; 35: 691–700.
14.
BesagJ. Spatial interaction and the statistical analysis of lattice systems (with discussion). J R Stat Soc B1974; 36: 192–225.
15.
RueHHeldL. Gaussian Markov random fields. Boca Raton, FL: Chapman and Hall/CRC, 2005.
16.
SørbyeSHRueH. Scaling intrinsic Gaussian Markov random field priors in spatial modelling. Spat Stat2014; 8: 39–51.
17.
Freni-SterrantinoAVentrucciMRueH. A note on intrinsic conditional autoregressive models for disconnected graphs. Spat Spatiotemporal Epidemiol2018; 26: 25–34.
18.
GelmanA. Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper). Bayesian Anal09 2006; 1: 515–534.
19.
FongYRueHWakefieldJ. Bayesian inference for generalized linear mixed models. Biostatistics2010; 11: 397–412.
20.
SimpsonDRueHRieblerA, et al. Penalising model component complexity: a principled, practical approach to constructing priors. Stat Sci02 2017; 32: 1–28.
21.
RieblerASørbyeSHSimpsonD, et al. An intuitive Bayesian spatial model for disease mapping that accounts for scaling. Stat Methods Med Res2016; 25: 1145–1165. PMID: 27566770.
22.
FuglstadG-AHemIGKnightA, et al. Intuitive joint priors for variance parameters. Bayesian Anal2020; 15: 1109–1137.
23.
VentrucciMCocchiDBurgazziG, et al. PC priors for residual correlation parameters in one-factor mixed models. Stat Methods Appl2020; 29: 745–765.
24.
BesagJYorkJMollieA. Bayesian image restoration, with two applications in spatial statistics. Ann Inst Stat Math1991; 43: 1–21.
25.
GoicoaTAdinAUgarteMD, et al. In spatio-temporal disease mapping models, identifiability constraints affect PQL and INLA results. Stoch Environ Res Risk Assess2018; 32: 749–770.
26.
FahrmeirLKneibTLangS, et al. Regression: Models, methods and applications. Berlin: springer-Verlag, 2013.
27.
WahbaG. Improper priors, spline smoothing and the problem of guarding against model errors in regression. J R Stat Soc B (Methodol)1978; 40: 364–372.
28.
Frühwirth-SchnatterSWagnerH. Stochastic model specification search for Gaussian and partial non-Gaussian state space models. J Econom2010; 154: 85–100.
29.
Frühwirth-SchnatterSWagnerH. Bayesian variable selection for random intercept modeling of Gaussian and non-Gaussian data. In: BernardoJMBayarriMJBergerJODawidAPHeckermanDSmithAFMWestM (eds) Bayesian statistics 9. Oxford, Oxford University Press, 2011, pp. 165–200.
30.
VentrucciMRueH. Penalized complexity priors for degrees of freedom in Bayesian p-splines. Stat Model2016; 16: 429–453.
31.
KullbackSLeiblerRA. On information and sufficiency. Ann Math Stat1951; 22: 79–86.
32.
RueHMartinoSChopinN. Approximate Bayesian inference for latent Gaussian models using integrated nested Laplace approximations (with discussion). J R Stat Soc B2009; 71: 319–392.
33.
SpiegelhalterDJBestNGCarlinBP, et al. Bayesian measures of model complexity and fit (with discussion). J R Stat Soc B2002; 64: 583–639.
34.
WatanabeS. A widely applicable Bayesian information criterion. J Mach Learn Res2013; 14: 867–897.
35.
SørbyeSHRueH. Fractional Gaussian noise: prior specification and model comparison. Environmetrics2018; 29: e2457.
36.
KassRERafteryAE. Bayes factors. J Am Stat Assoc1995; 90: 773–795.
37.
RieblerAHeldL. The analysis of heterogeneous time trends in multivariate age-period-cohort models. Biostatistics2010; 11: 57–69.
38.
PicadoAGuitianJPfeifferD. Space–time interaction as an indicator of local spread during the 2001 FMD outbreak in the UK. Prev Vet Med05 2007; 79: 3–19.
39.
GelmanA. Analysis of variance-why it is more important than ever. Ann Stat2005; 33: 1–53.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
