Comments on: Allowing intercepts and slopes to vary in a logistic regression: how does this change the ROC curve?
https://statmodeling.stat.columbia.edu/2019/08/03/allowing-intercepts-and-slopes-to-vary-in-a-logistic-regression-how-does-this-change-the-roc-curve/
Mon, 05 Aug 2019 16:48:43 +0000
hourly
1 https://wordpress.org/?v=5.4.2
By: John Pugliese
https://statmodeling.stat.columbia.edu/2019/08/03/allowing-intercepts-and-slopes-to-vary-in-a-logistic-regression-how-does-this-change-the-roc-curve/#comment-1097098
Mon, 05 Aug 2019 16:48:43 +0000https://statmodeling.stat.columbia.edu/?p=41303#comment-1097098They might consider a latent-class analysis to identify the groups and follow that with LR on the outcome of interest.
]]>
By: sam
https://statmodeling.stat.columbia.edu/2019/08/03/allowing-intercepts-and-slopes-to-vary-in-a-logistic-regression-how-does-this-change-the-roc-curve/#comment-1096884
Mon, 05 Aug 2019 02:03:09 +0000https://statmodeling.stat.columbia.edu/?p=41303#comment-1096884I’ve thought about theoretical results but I think you’d have to make too many assumptions for it to be at all useful. I think of AUC as a measure of how overlapping two distributions are. If there is no overlap in the distributions then AUC is 1 and if there is total overlap AUC is .5. So first assumption you’d have to make is the distributions of both distributions. Suppose they are normal(0,1) and normal(1,1) for instance; then you’d get an AUC of ~.77. Better models decrease the standard deviation or reduce the difference in means of the distributions so you’d have to assume how much better your new data would be. So you have to 1) assume distributions to classes and 2) have an expected increase in predictive power with new features. But there’s no theoretical way of determining either without trying it out…
]]>
By: C. Luke Watson
https://statmodeling.stat.columbia.edu/2019/08/03/allowing-intercepts-and-slopes-to-vary-in-a-logistic-regression-how-does-this-change-the-roc-curve/#comment-1096831
Sun, 04 Aug 2019 20:09:36 +0000https://statmodeling.stat.columbia.edu/?p=41303#comment-1096831This reminds me of a finite mixture model. In which case, a GMM approach is the following:(Fox, Kim, Ryan, Bajari; 2011) http://fox.web.rice.edu/published-papers/fox-kim-ryan-bajari-qe.pdf