The problem of variable selection is encountered in model fitting with unobserved spatial predictors at the sites where outcomes are measured. The variability of the interpolated predictors at outcome sites results in potential problems of variable selection and averaging the results across different datasets. A Bayesian spatial model is developed to tackle this issue. By sampling the latent spatial predictors and selecting the spatial and non-spatial predictors using stochastic search variable selection Gibbs sampling algorithm, our approach allows for uncertainty of the predictors including the interpolated predictors. The approach is evaluated and illustrated through a simulated data example and an application to mental retardation and developmental delay in a Medicaid population in South Carolina with samples of soil chemistry.
BanerjeeSCarlinBPGelfrandAECarlinBP (2004) Hierarchical modeling and analysis for spatial data. Boca Raton, FL: Chapman & Hall/CRC Press.
2.
BeelingerDC (2006) Human developmental neurotoxicology. New York: Taylor & Francis.
3.
CarlinBPChibS (1995) Bayesian model choice via Markov chain Monte Carlo methods. Journal of the Royal Statistical Society Series B, 57, 473–84.
4.
CarrollRRuppertDStefanskiLCrainiceanuC (2006) Measurement error in nonlinear models: a modern perspective. Boca Raton, FL: Chapman & Hall/CRC.
5.
CressieNAC (1993) Statistics for spatial data, revised edition, Wiley.
6.
CuiWGeorgeEI (2008) Empirical Bayes vs. fully Bayes variable selection. Journal of Statistical Planning and Inference, 138, 888–900.
7.
DellaportasPForsterJJNtzoufrasI (2002) On Bayesian model and variable selection using MCMC. Statistics and Computing, 12, 27–36.
8.
DenisonDGTHolmesCCMallickBKSmithAFM (2002) Bayesian methods for nonlinear classification and regression. Chichester: John Wiley and Sons.
9.
FinleyAOBanerjeeSCarlinBP (2007) spBayes: An R package for univariate and multivariate hierarchical point-referenced spatial models. Journal of Statistical Software, 19, 4.
10.
FrenchJLWandMP (2004) Generalized additive models for cancer mapping with incomplete covariates. Biostatistics, 5, 177–91.
11.
FuentesMSongHRGhoshSHollandDDavisJ (2006) Spatial association between speciated particulate matter and mortality. Biometrics, 62, 855–63.
12.
GeisserS (1984) On prior distribution for binary trials. American Statistician, 38, 244–51 (with discussion).
13.
GeorgeEIMcCullochRE (1993) Variable selection via Gibbs sampling. Journal of the American Statistical Association, 88, 881–89.
14.
GewekeJ (1992) Evaluating the accuracy of sampling-based approaches to calculating posterior moments. In BernadoJMBergerJODawidAPSmithAFM (eds), Bayesian statistics. Oxford, UK: Clarendon Press.
15.
GilksWRBestNGTanKKC (1995) Adaptive rejection Metropolis sampling within Gibbs sampling. Journal of Applied Statistics44, 455–72.
16.
GoldmanLRKoduruS (2000) Chemicals in the environment and developmental toxicity in children: a public health and policy perspective. Environmental Health Perspectives108(S3), 443–48.
17.
GreenPJ (1995) Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika, 82, 711–32.
18.
HutchinsonMFGesslerPE (1994) Splines—more than just a smooth interpolator. Geoderma, 62, 45–67.
19.
KammannEEWandMP (2003) Geoadditive models. Journal of the Royal Statistical Society Series C, 52, 1–18.
20.
KimJLawsonABMcDermottSAelionCM (2010) Bayesian spatial modeling of disease risk in relation to multivariate environmental risk fields. Statistics in Medicine, 29, 142–57.
LaslettGM (1994) Kriging and splines: an empirical comparison of their predictive performance in some applications. Journal of the American Statistical Association, 89, 391–400.
23.
LaslettGMMcBratneyAB (1990) Further comparison of spatial methods for predicting soil pH. Soil Science Society of America Journal, 54, 1553–58.
LeyESteelMF (2009) On the effect of prior assumptions in Bayesian model averaging with applications to growth regression. Journal of Applied Econometrics24, 651–74.
26.
LittleRJARubinDB (2002) Statistical analysis with missing data, 2nd edition.New York: John Wiley.
27.
LunnDThomasABestNSpiegelhalterD (2000) WinBUGS: A Bayesian modeling framework: Concepts, structure, and extensibility. Statistics and Computing, 10, 325–37.
NychkaDHaalandPO’ConnellMEllnerS (1998) FUNFITS, data analysis and statistical tools for estimating functions. In NychkaDPiegorschWWCoxLH (eds), 159–79. Case Studies in Environmental Statistics. New York: Springer.
30.
R Development Core Team (2007) R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. http://www.R-project.org.
31.
RuppertDWandMPCarrollRJ (2003) Semiparametric regression. Cambridge: Cambridge University Press.
32.
ScottJBergerJ (2008) Multiple testing, empirical Bayes, and the variable-selection problem. Discussion Paper 2008-10: Duke University Department of Statistical Science.
33.
SullivanJBKriegerGR (2001) Clinical environmental health and toxic exposures, 2nd edition.Philadelphia, PA: Lippincott Williams & Wilkins.
34.
VoltzMWebsterR (1990) A comparison of kriging, cubic splines and classification for predicting soil properties from sample information. European Journal of Soil Science, 41, 473–90.
35.
WoodS (2006) Generalized additive models: an introduction with R, CRC Press.
