Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Output summary #17

Closed
TaylorOshan opened this issue Jun 15, 2018 · 9 comments
Closed

Output summary #17

TaylorOshan opened this issue Jun 15, 2018 · 9 comments

Comments

@TaylorOshan
Copy link
Collaborator

It would be good to have a summary function that constructed a text statement with a summary of the GWR/MGWR results that could be printed or saves to file, such as is done in GWR4/GWmodel.

@weikang9009
Copy link
Member

I think I have one when I was writing the MGWR GUI a long time ago. Let me dig it.

@TaylorOshan
Copy link
Collaborator Author

That would be great! Could be good to mention and highlight it in the JSS manuscript if we have it updated for both gwr/mgwr.

@Ziqi-Li
Copy link
Member

Ziqi-Li commented Jun 20, 2018

Hi
I have previously made one summary function for the GUI, which is based on Wei's summary code and compatible with our current APIs.

It looks like this.

But two things not sure how to deal with:

  1. Do we need to output summary for global regression? It makes sense to me to put the summary function into GWRResults class as GWRResults.summary. And maybe only output GWR results?

  2. Since our interface takes numpy array, we do not have variable names, and they may be needed in the summary. Any solution?

@TaylorOshan
Copy link
Collaborator Author

On 1, I think it would be good to have. Could utilize spglm, since its already a dependency and include with the output from GWRResults.summary.

On 2, I think it makes sense to have a field_name parameter that accepts a list of names either in the GWR class or to GWRResults.summary method. Could actually pass it to either of those two and use it if it is available. If not, could auto-generate placeholder names like x1, x2, x3, etc.

@Ziqi-Li
Copy link
Member

Ziqi-Li commented Jun 21, 2018

Shipped with #18.
Hi @TaylorOshan, do you mind adding the aicc for global glm in spglm? I could only find aic and bic but not aicc. It would be easier if you could do a quick PR to spglm, I'm less familiar with that lib. It would be good to have the aicc in global regression summary output.

@TaylorOshan
Copy link
Collaborator Author

I'm a bit apprehensive to add it to spglm because I don't think it is used for GLM's outside of the GWR framework. Perhaps we could generalize the AICc in diagnostics.py to also take a GLMResults class by adding a tr_S attribute?. Then the AICc could be computed on a global and GWR/MGWR models.

@Ziqi-Li
Copy link
Member

Ziqi-Li commented Jun 22, 2018

Sure, if you could do that, I will push a new commit to include aicc for global model.

@TaylorOshan
Copy link
Collaborator Author

Done in spglm/#7, should allow aicc to be calculated for GLMResults object.

@TaylorOshan
Copy link
Collaborator Author

Satisfied by #18

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants