bigDM

Scalable Bayesian disease mapping models for high-dimensional data using a divide and conquer approach.

The package
Installation
Basic Use
Updates
Copyright and license

The package

This package implements several (scalable) spatial and spatio-temporal generalised linear mixed models for high-dimensional areal count data in a fully Bayesian setting using the integrated nested Laplace approximation (INLA) technique.

Below, there is a list with a brief overview of all package functions:

add_neighbour Adds isolated areas (polygons) to its nearest neighbour.
CAR_INLA Fits several spatial CAR models for high-dimensional count data.
clustering_partition Obtain a spatial partition using the DBSC algorithm.
connect_subgraphs Merges disjoint connected subgraphs.
divide_carto Divides the spatial domain into subregions.
mergeINLA Merges inla objects for partition models.
random_partition Defines a random partition of the spatial domain based on a regular grid.
STCAR_INLA Fits several spatio-temporal CAR models for high-dimensional count data.

Installation

Installing Rtools40 for Windows

R version 4.0.0 and newer for Windows requires the new Rtools40 to build R packages with C/C++/Fortran code from source.

Install from CRAN

install.packages("bigDM")

Install from GitHub (development version)

# Install devtools package from CRAN repository
install.packages("devtools")

# Load devtools library
library(devtools)

# Install the R-INLA package
install.packages("INLA", repos=c(getOption("repos"), INLA="https://inla.r-inla-download.org/R/stable"), dep=TRUE)

# In some Linux OS, it might be necessary to first install the following packages
install.packages(c("cpp11","proxy","progress","tzdb","vroom"))

# Install bigDM from GitHub repositoy
install_github("spatialstatisticsupna/bigDM")

IMPORTANT NOTE: At least the stable version of INLA 22.11.22 (or newest) must be installed for the correct use of the bigDM package.

Basic Use

See the following vignettes for further details and examples using this package: * bigDM: fitting spatial models * bigDM: parallel and distributed modelling * bigDM: fitting spatio-temporal models

When using this package, please cite the following papers:

Orozco-Acosta, E., Adin, A., and Ugarte, M.D. (2021). Scalable Bayesian modeling for smoothing disease risks in large spatial data sets using INLA. Spatial Statistics, 41, 100496.

Orozco-Acosta, E., Adin, A., and Ugarte, M.D. (2022). Parallel and distributed Bayesian modelling for analysing high-dimensional spatio-temporal count data. arXiv preprint.

Updates

news(package="bigDM")

Changes in version 0.4.2 (2022 Jun 27) * small bugs fixed * new merging strategy

Changes in version 0.4.1 (2022 Feb 01) * small bugs fixed * version submmited to CRAN

Changes in version 0.4.0 (2022 Jan 21) * new STCAR_INLA() function to fit scalable spatio-temporal CAR models

Changes in version 0.3.2 (2021 Nov 05) * X and confounding arguments included to CAR_INLA() function * new function included: clustering_partition()

Changes in version 0.3.1 (2021 May 03) * W argument included to CAR_INLA() function

Changes in version 0.3.0 (2021 Apr 19) * parallel and distributed computation strategies when fitting inla models with the CAR_INLA() function

Changes in version 0.2.2 (2021 Mar 12) * new arguments included to random_partition() function

Changes in version 0.2.1 (2021 Feb 25) * Carto_SpainMUN data changed

Changes in version 0.2.0 (2020 Oct 01) * speedup improvements in mergeINLA() function * small bugs fixed

Acknowledgments

This work has been supported by Project MTM2017-82553-R (AEI/FEDER, UE) and Project PID2020-113125RB-I00/MCIN/AEI/10.13039/501100011033. It has also been partially funded by the Public University of Navarra (project PJUPNA2001) and by la Caixa Foundation (ID 1000010434), Caja Navarra Foundation and UNED Pamplona, under agreement LCF/PR/PR15/51100007 (project REF P/13/20).