Skip to content

R packages for multivariate spatial models

6 messages · Sarah Goslee, Jin Li, Roger Bivand +1 more

#
Dear list members,

I am looking for suggestions of R packages to fit multivariate spatial
models.
#
Hi,

That's a huge topic! You might start by going to rseek.org and
searching for multivariate spatial models. There are many packages
that perform that kind of analysis.

Browsing the Spatial task view at
https://cran.r-project.org/web/views/ and maybe also the Multivariate
task view might give you a broader view of the multivariate and
spatial capabilities of R packages.

If you can't find what you are looking for, a more specific question
might get you more specific answers. :)

Sarah
On Sun, Apr 7, 2024 at 5:14?PM Manuel Sp?nola <mspinola10 at gmail.com> wrote:

  
    
#
Hi Manuel,

If your models are for predictive purpose, `spm` and `spm2` are perhaps the
right packages for you. Some methods for spatial predictive modelling are
introduced in the `spm` and `spm2` packages, including such as modern
statistical methods (i.e., generalised linear models, glmnet, generalised
least squares), random forest, gbm, svm, kriging methods, and a number of
hybrid methods. For more information, please see "Spatial Predictive
Modeling with R" that provides many reproducible examples of these methods.

Hope this helps,
On Mon, Apr 8, 2024 at 7:14?AM Manuel Sp?nola <mspinola10 at gmail.com> wrote:

            

  
    
#
Thank you very much Sarah and Jin,

I am referring to multivariate response spatial models.



El dom, 7 abr 2024 a las 22:52, Jin Li (<jinli68 at gmail.com>) escribi?:

  
    
#
Is https://doi.org/10.32614/RJ-2023-068 more relevant? The article has suppleementary materials including multivariate responses for a number of settings.

Hope this helps,

Roger

--
Roger Bivand
Emeritus Professor
Norwegian School of Economics
Postboks 3490 Ytre Sandviken, 5045 Bergen, Norway
Roger.Bivand at nhh.no
#
Thank you very much Roge.

Yes, this is the kind of multivariate spatial model I am looking for.
Particularly for areal or lattice data.



El lun, 8 abr 2024 a las 6:08, Roger Bivand (<Roger.Bivand at nhh.no>)
escribi?: