How do I choose between these models?
When choosing between these various types of models, it is important to understand your data as well as the assumptions that go into each of the various models. It is also important to balance model performance with your individual operational constraints, such as how long you’re willing to wait for model results. Here are a few questions that you can ask yourself to help determine which model may be better for each situation:
- Do the patterns between my target and explanatory variables vary across space?
- If the answer to this question is yes, then fitting a GWR or an MGWR model may be a better-suited option, as OLS fits a global regression compared to the local regression fit by GWR and MGWR.
- If the patterns between my target and explanatory variables vary across space, do they also operate at different scales?
- If the answer to these questions is yes, then MGWR is a better-suited option than GWR. Recall that GWR assumes a single...
