Timeline for Testing for feature importance with missing values
Current License: CC BY-SA 3.0
2 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 17, 2016 at 17:11 | comment | added | James Eaves | This line of thought is helpful. Focusing on one period, like the first semester, reduces the missing observations caused by dropouts. But what I find is that there is more variability, than expected, in the courses students take (even within the same major). So I could just consider courses that all (or most) students took - these are core courses, so there is a logic to that. Though since students may decide to take these courses at different semesters, this may introduce self-selection bias - but that choice may also be useful information. Thanks- I'd up-vote your response if I could :) | |
| Jun 16, 2016 at 20:22 | history | answered | geekoverdose | CC BY-SA 3.0 |