Hi, I have a problem to select the best model between spatial panel (SAR and SEM) and nonspatial panel. The results of LM lag and LM error show insignificant value, but rho/lamda is significant. Does it means spatial model is not necessary? Because most of literature refer to LM tests, not rho/lambda coefficient.
Thanks.
The GeoDa documentation discusses the role of rho and lambda.
https://s4.ad.brown.edu/resources/tutorial/modul2/geoda3final.pdf
http://labs.bio.unc.edu/buckley/documents/anselinintrospatregres.pdf
The insignificant LM tests on your SAR and SEM models indicate that there is no further spatial autocorrelation (SA) in these models. The significant rho and lamda values suggest that there was SA in the error terms of the OLS non-spatial model. To test this, you should use the Moran I test on the error terms of your OLS model. I would expect that it is significant. Your next hurdle is to choose between the SAR and SEM model. In principle, this should be based on theoretical arguments. For instance, is there a physical or economic justification for the dependent variable to be spatially autocorrelated? If there is not, the SEM model is preferable. You might also test the GSM (General Spatial Model), that combines the SAR and SEM.
Correct me if I'm wrong please, but the way I understand it:
When we run 'spatdiag' command, we get both LM and robust LM for SAR/SEM:
to decide which model is preferred:
1) first compare the LM SEM vs LM SAR, whichever significant, that is the preferred model.
2) if both are significant, then compare the robust LM, whichever significant, then that is the preferred model.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix