Supplementary material for
Paciorek, C.J. 2013. Spatial models for point and areal data using
Markov random fields on a fine grid. Electronic Journal of Statistics.

Contact information:
Christopher Paciorek
paciorek@alumni.cmu.edu

Files in the archive:

ejsCode.R: primary code for all analyses, simulations, examples, figures 
ejsAuxil.R: a few auxiliary functions used in ejsCode.R
graph{Bin,Tps}AxB.csv: adjacency information for the ICAR and TPS-MRF models for grids of size AxB for 75x75, 100x100, 99x122
graphCatsTpsAxB.csv: additional adjacency information for the TPS-MRF model
gridForPM.csv: latent grid for the PM example
lacounty00tr_intgrid.csv: info on the intersections between census tracts and grid cells for the LA cancer example
LAgrid.csv: latent grid for LA example
LAwithoutData.RData: R Data file containing census tract info but without the actual cancer counts as these are not publicly available
pm.csv: data for the PM example
pmFitMod{,2,3}.txt: fitted values for the three LA cancer model fits plotted in Fig. 10
Qobjects.RData: R Data file containing the spam format sparse precision matrices for the (large) grid used in the PM example
uscontermprj.*: shapefile files for the geographic projection of the northeast US used in the PM example