GeoAdjust: Adjusting for Positional Uncertainty in Geostatistial Analysis of DHS Data

The R-package GeoAdjust adjusts for positional uncertainty in GPS coordinates and performs fast empirical Bayesian geostatistical inference for household survey data from the Demographic and Health Surveys (DHS) Program. DHS household survey data is important for tracking demographic and health indicators, but is published with intentional positional error to preserve the privacy of the household respondents. Such jittering has recently been shown to deteriorate geostatistical inference and prediction, and GeoAdjust is the first software package that corrects for jittering in geostatistical models containing both spatial random effects and raster- and distance-based covariates. The package provides inference for model parameters and predictions at unobserved locations, and supports Gaussian, binomial and Poisson likelihoods with identity link, logit link, and log link functions, respectively. GeoAdjust provides functions that make model and prior specification intuitive and flexible for the user, as well as routines for plotting and output analysis.