I have read that the following 6 criteria need to be met for the OLS model with respect to variables prior to performing a GWR on them: (Is this correct?) and ...

1) Spatial Autocorrelation (my understanding is that spatial auto correlation is acceptable for GWR, since the whole reason for using GWR is to look at local patterns

2) Checking coefficient signs and ensuring that there is a negative for expected inverse relationships

3) Checking redundancy between variables using the VIF (should be less than <7.5)

4) Ensuring variables are statistically significant (p<0.05) and removing ones that are not prior to GWR

5) Ensuring that the residuals of over/under predictions are the result of random noise using the Jarque–Bera (JBT) test, this test result should not be significant

6) Looking at the adjusted R2, a good model is > 0.5

My JBT test is significant, what does this mean in terms of GWR? Can I still proceed or should I transform my data? Use a different model?

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.