Timeline for Comparing probabilities of two models
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 10, 2024 at 13:30 | history | edited | Ben Reiniger♦ | CC BY-SA 4.0 | remove first part that misinterpreted the question, address it in the second part |
| Dec 9, 2024 at 22:22 | comment | added | Ben Reiniger♦ | You've got it on the second one. There are degrees of calibration; GBMs tend to be overconfident, pushing predicted probabilities toward 0 and 1, but sometimes they're close enough that I think the comparisons you're after would be OK. | |
| Dec 9, 2024 at 21:58 | comment | added | Ale | Thank you for your answer but I'm not sure I got it. In the first setting, is it a yes? To clarify, I'm considering two different models, each one applied on a different binary output of the same training dataset, not a single multiclass model. Regarding the second answer, I understand that the claim holds only if the results are "well-calibrated", which is something I will search on, and that maybe is not even feasible with gradient boosted trees. Is that it? | |
| Dec 9, 2024 at 16:55 | history | answered | Ben Reiniger♦ | CC BY-SA 4.0 |