0

I am looking to create a predictive model in ArcGIS. I have done this previously with glm with poisson distribution and found it relatively straightforward in raster calculator e.g.

exp(intercept + (xvariable) + (yvariable))

However, it is likely that i shall need to use a more complex approach to account for both local and landscape variables and will need to use a Generalised Linear Mixed Model.

Does anyone have experience of this? Is it possible to do in ArcGIS(raster calculator), how would the equation likely differ to that above?

1 Answer 1

1

You might be able to create a python geoprocessing script for ArcGIS, and load in some python libraries to do the stats for you. There are a lot of stats libraries out there, a good place to start would be:

http://statsmodels.sourceforge.net/devel/glm.html

I also remember reading about how you can create R scripts within python as well. A user would need to have both ArcGIS and R installed and licensed on their system to use it. I haven't done it yet, but am seriously thinking about trying to use R with ArcGIS in the near future, so I'd be curious how it goes if you go that route!

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.