[R-sig-Geo] spatial anova

[Dexter Locke]

Normally I would agree with Steve, but this is a topic I'm interested in
and have searched extensively. I have not yet found a useable
implementation. If there is some R code that can do this, I would love to
hear about it.

In April Babak Naimi <naimi.b at gmail.com> posted on this list a very similar
question, but I did not see any replies.

For the underlying maths, see these papers:

Griffith, D. A. (1978). A spatially adjusted ANOVA model. *Geographical
Analysis*, *X*(3), 298–301. doi:10.1016/0166-0462(92)90034-X

Griffith, D. A. (1992). A spatially adjusted N-way ANOVA model. *Regional
Science and Urban Economics*, *22*(3), 347–369.
doi:10.1016/0166-0462(92)90034-X

These are also related:

Long, J. a., Nelson, T. a., & Wulder, M. a. (2010). Local indicators for
categorical data: Impacts of scaling decisions. *Canadian Geographer*, *54*(1),
15–28. doi:10.1111/j.1541-0064.2009.00300.x

Pietrzak, M. B., Wilk, J., Kossowski, T., & Bivand, R. (2014). The
application of local indicators for categorical data (LICD) in the spatial
analysis of economic development, (14). Retrieved from
http://ideas.repec.org/p/pes/wpaper/2014no14.html

The old SpaceStat used to have a spatial anova option (white paper by
Anselin):

http://siteresources.worldbank.org/INTPGI/Resources/342674-1092157888460/Anselin.spacestatTutorial.pdf

SpaceStat 4 appears to be able to perform a spatial anova:

https://www.biomedware.com/files/SpaceStat_4.0_Documentation.pdf


Just to be sure I hadn't missed anything I double checked and found this:

library(spdep); help(joincount.multi) and joincount.test(). So it looks
like Roger Bivand provided the needed code once again! Thanks so much.

HTH,
Dexter




On Tue, Jun 28, 2016 at 12:20 PM, Steve Friedman <friedman.steve at gmail.com>
wrote:

> I would start by reading google and R posts regarding spatial anova. There
> is plenty of help among that literature.  Formulate your model, share it,
> and then you might have better questions.
>
> Steve
>
> On Tuesday, June 28, 2016, Csany22 csany <csany22 at hotmail.com> wrote:
>
> > Hi R-sig-geo people,
> >
> >
> > what would be the best way to run spatial anova in R? I have a dependent
> > (continuous), independent (categorical) variables and lat and long fo
> reach
> > of the cell(pixel) based observations. I need to run anova. as would like
> > to compare the means (Tukey) of the independent variable.
> >
> >         [[alternative HTML version deleted]]
> >
> > _______________________________________________
> > R-sig-Geo mailing list
> > R-sig-Geo at r-project.org <javascript:;>
> > https://stat.ethz.ch/mailman/listinfo/r-sig-geo
> >
>
>
> --
> Steve Friedman
> Cell: 517-648-6290
> Web:  www.envisioningecosystemsdecisions.com
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

	[[alternative HTML version deleted]]