Spatial autocorrelation

How can I apply GLM and GAM to spatially autocorrelated data?

I need to apply generalized linear modeling and additive modeling to my data (continuous and categorical explanatory variables). But my data is spatially autocorrelated. What is the best options?

1. you could subsample your data until the Moran's I (or variogram) indicates that there is no longer spatial correlation.
2. you could group the data so that locations within a certain distance of each other belong to a group variable. You would then specify the intercept of your models as a random effect across groups. not sure how this would work… DAC
3. Or you could model the spatial correlation explicitly. Take a look at this website. It describes things fairly well and has example code... http://rstudio-pubs-static.s3.amazonaws.com/9687_cc323b60e5d542449563ff1142163f05.html

There is a review here: "Methods to account for spatial autocorrelation in the analysis of species distributional data: a review" by Dormann et al (2007). http://www2.unil.ch/biomapper/Download/Dormann-EcoGra-2007.pdf

There’s a Bayasian approach in R, based on GAMS You can also have a look at this package in R

https://people.maths.bris.ac.uk/~sw15190/

This site is a nice explanation of how to model autocorrelation in nlme
http://eco-stats.blogspot.co.uk/2014/10/r-lab-inference-with-spatially.html