Timeline for Do CART trees capture interactions among predictors?
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 21, 2020 at 8:02 | history | edited | Antoine | CC BY-SA 4.0 |
added 420 characters in body
|
| May 6, 2019 at 17:08 | answer | added | Anton Tarasenko | timeline score: 8 | |
| Apr 22, 2015 at 7:01 | comment | added | Antoine | @mbq so the new splits are conditional with respect to the preceding splits... I see... I guess I was having trouble understanding that "conditioned by a previous split made on a given predictor" was equivalent to "interacting with this predictor"... | |
| Apr 21, 2015 at 23:50 | history | edited | user88 | CC BY-SA 3.0 |
edited title
|
| Apr 21, 2015 at 23:50 | comment | added | user88 | The flaw in the argument is that splits are made on subsets of covariates defined by splits done previously. | |
| Apr 21, 2015 at 21:13 | vote | accept | Antoine | ||
| Apr 21, 2015 at 21:08 | history | tweeted | twitter.com/#!/StackStats/status/590623062757892096 | ||
| Apr 21, 2015 at 20:34 | answer | added | TrynnaDoStat | timeline score: 20 | |
| Apr 21, 2015 at 19:30 | history | asked | Antoine | CC BY-SA 3.0 |