Reinier van Linschoten
Reinier van Linschoten
@al-cheb, see below (zip because I can't upload .cap files) [sntp capture.zip](https://github.com/actions/runner-images/files/9335952/sntp.capture.zip)
I will do this, but I see that some API specifications changed, so I'll need to adjust my testing suite/code first. I'll get back to you with the proper capture...
Yes, it seems that with a change in backend architecture for the API, the tests are less flaky. Closing this issue for now.
The issue is not in DHARMa, but in the predict function of glmmTMB, see my comment in the other issue. I think @florianhartig can close this as the problem is...
Awesome, thank you very much for fixing this!
> Other than the issue reported by Daniel, `icc()` returns the model implied ICC on the link scale, so it is generally expected the use of different families or link...
That is an interesting option, but I am looking for the ICC (adjusted) for one of the grouping factors (random intercepts) in my data. Is there a comparable workaround?
> Are you allowed to share the data? This would help in debugging. No, unfortunately the data is very sensitive and we are under no circumstance allowed to share it....
Hi all, No, was not able to fix it unfortunately. I calculated the calibration intercept and slope, with corresponding CIs myself. It's quite easy actually. ```r # First calculate logits...
Great! C-statistic (also called AUC) with CI can be done with the pROC package Op di 12 mrt 2024 21:17 schreef RRiviere15 ***@***.***>: > Hi Renier, > > That's brilliant....