mgwrsar: GWR, Mixed GWR with Spatial Autocorrelation and Multiscale GWR (Top-Down Scale Approaches)

Provides methods for Geographically Weighted Regression with spatial autocorrelation (Geniaux and Martinetti 2017) <doi:10.1016/j.regsciurbeco.2017.04.001>. Implements Multiscale Geographically Weighted Regression with Top-Down Scale approaches (Geniaux 2026) <doi:10.1007/s10109-025-00481-4>.

Version:	1.3.1
Depends:	R (≥ 3.5.0), sp, Matrix
Imports:	Rcpp, ggplot2, sf, knitr, methods, doParallel, foreach, nabor, mapview, rlang, dplyr, gridExtra, grid, mboost, mgcv, caret, stringr, SMUT, plotly, RhpcBLASctl, magrittr, lifecycle
LinkingTo:	Rcpp, RcppEigen, RcppArmadillo
Suggests:	rmarkdown
Published:	2026-01-21
DOI:	10.32614/CRAN.package.mgwrsar
Author:	Ghislain Geniaux [aut, cre], Davide Martinetti [aut], César Martinez [aut]
Maintainer:	Ghislain Geniaux <ghislain.geniaux at inrae.fr>
License:	GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation:	yes
SystemRequirements:	C++17
Citation:	mgwrsar citation info
Materials:	NEWS
CRAN checks:	mgwrsar results

Documentation:

Reference manual:

mgwrsar.html , mgwrsar.pdf

Vignettes:

GWR and Mixed GWR with spatial autocorrelation (source, R code) GWR and MGWR with Space-Time Kernels (source, R code) Estimating GWR and Mixed GWR Models with mgwrsar package: An Introduction with House Price Data (source, R code) Multiscale GWR using top down scale approaches (source, R code) Speeding up GWR like models with mgwrsar package using Target Points, rough gaussian kernel and parallelisation (source, R code)

Downloads:

Package source:	mgwrsar_1.3.1.tar.gz
Windows binaries:	r-devel: mgwrsar_1.1.zip, r-release: mgwrsar_1.1.zip, r-oldrel: mgwrsar_1.1.zip
macOS binaries:	r-release (arm64): mgwrsar_1.1.tgz, r-oldrel (arm64): mgwrsar_1.1.tgz, r-release (x86_64): mgwrsar_1.1.tgz, r-oldrel (x86_64): mgwrsar_1.1.tgz
Old sources:	mgwrsar archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=mgwrsar to link to this page.