My knowledge of this topic is rather limited. But essentially i have a linear model based upon a number of predictor variables which i am running to produce an estimate.

I need to account for spatial autocorrelation within my data.

Am i understanding this correctly in that i can run my linear model through a GEE, without specifying coordinates for my data points and account for spatial auto correlation by specifying the distribution of the 'unknown'(spatial) correlation?

I have attempted other methods which account for spatial autocorrelation by processing the data with e.g. XCOR YCOR variables but R crashes. My sample size is over 2000.



Your Answer

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

Browse other questions tagged or ask your own question.