Are the Lagrange values comparable in lm.LMtests?

If both RLMerr and RLMlag are significant, would the test with the largest Lagrange value be considered 'better' at explaining the spatial relationship for the given model? Or 'better' at explaining which type of spatial autocorrelation is most prevalent in the model?

This is my first question, so apologies for probably not following the necessary protocol/formatting/salutation OR if this belongs in some other silo of StackExchange.


According to Anselin in this manual (page 12):

the first thing is to assess which of the LMerr and LMlag have significant p-values. If both are significant, then the next step is to compare thep-values of the robust forms RLMerr and RLMlag.

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.