Timeline for Methods to derive cut-offs for continuous variables
Current License: CC BY-SA 4.0
28 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 25 at 19:33 | vote | accept | Kate | ||
| Jan 25 at 7:38 | comment | added | Kate | @FrankHarrell sorry for not being clear in the post - I pass continuous variable to fit random forest and to identify the important variables. Once important variables are identified, I would like to identify the cut-off for continuous variables since it is valuable for clinicians. So I am not applying the cutoff before fitting random forest | |
| Jan 23 at 16:01 | history | edited | kjetil b halvorsen♦ | CC BY-SA 4.0 |
edited tags
|
| Jan 23 at 15:53 | answer | added | EdM | timeline score: 5 | |
| Jan 23 at 12:36 | comment | added | Frank Harrell | As discussed extensively on this site, cutoffs are bad ideas. They don’t exist in nature because discontinuities don’t exist in nature unless X=time. Since they don’t exist, every analyst will find a different cutoff. | |
| Jan 23 at 9:26 | history | edited | Kate | CC BY-SA 4.0 |
added more information in the post
|
| Jan 23 at 9:20 | history | edited | Kate |
changed tag as requested
|
|
| Jan 20 at 15:04 | history | reopened |
Pere Dave mdewey |
||
| Jan 20 at 13:58 | comment | added | Pere | I'm not very familiar with tags (nor with the topic of the question) but I would replace the r tag with classification tag. | |
| Jan 20 at 13:34 | comment | added | Kate | @Pere I am using conditional random forest and permimp::permimp for determining the importance of variables for my subgroup analysis . Now, I want to find cut-offs for continuous variables. But there are so many methods so I got lost | |
| Jan 20 at 13:31 | comment | added | Kate | @Dave which tag shall I add? Anything except the tag? | |
| Jan 20 at 10:43 | comment | added | Dave | The fact that this only has the r tag makes it seem that the question is just looking for some code, which remains off-topic here (not an inherently bad question, just outside the purview of Cross Validated). Perhaps this should be reopened to allow for a statistical response, however, perhaps even one that disputes the utility of the hard cutoffs the OP seems to seek. | |
| Jan 20 at 10:09 | review | Reopen votes | |||
| Jan 20 at 15:04 | |||||
| Jan 20 at 10:09 | comment | added | Pere | I don't think the question it's unclear but the OP can't be more specific for lack of specific knowledge. A good answer can pinpoint to the tools used to classify using cut-offs (or sort of), starting linear classification and continuing with random forest. statlearning.com may be a good source to start digging. | |
| Jan 19 at 14:35 | history | left closed in review |
Adrian Keister utobi whuber♦ |
Original close reason(s) were not resolved | |
| Jan 19 at 14:27 | comment | added | Kate | Let us continue this discussion in chat. | |
| S Jan 19 at 14:24 | review | Reopen votes | |||
| Jan 19 at 14:35 | |||||
| S Jan 19 at 14:24 | history | edited | Kate | CC BY-SA 4.0 |
updated the question
Added to review
|
| Jan 19 at 12:42 | comment | added | Dave | Multiple trusted moderators of this community disagree and have put the question on hold (“closed”) until such clarification comes. | |
| Jan 19 at 11:40 | comment | added | Kate | @Dave I already have it in my post | |
| Jan 19 at 11:10 | comment | added | Dave | Once you clarify the statistical content of your question, we’re all ears. | |
| Jan 19 at 11:05 | comment | added | Kate | @Dave I was asking about other methods if you would have read my question properly | |
| Jan 19 at 10:58 | history | closed |
Peter Flom PBulls User1865345 |
Not suitable for this site | |
| Jan 19 at 10:52 | comment | added | Dave | What about when a feature has more than one split to fit nonlinear behavior? (I think that can happen.) $//$ Maybe it would be best to ask about your ultimate goal once you get these splits. Extracting the splits is a programming question that is considered off-topic here, but there might be a real statistics question beneath it. | |
| Jan 19 at 10:47 | review | Close votes | |||
| Jan 19 at 10:58 | |||||
| Jan 19 at 10:46 | comment | added | Kate | @Dave something like the average/median cut-off used within forest? | |
| Jan 19 at 10:35 | comment | added | Dave | Wouldn’t there be a different splitting value (cutoff) in every tree in the random forest? | |
| Jan 19 at 10:29 | history | asked | Kate | CC BY-SA 4.0 |