Timeline for Multiple Comparison Problems
Current License: CC BY-SA 3.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 19, 2017 at 17:41 | comment | added | Moony | I guess that could be what she meant: in other words, maybe the fact that i am running an unadjusted model then trying different combinations of covariates would lead to an issue with multiple comparisons? | |
| Jan 19, 2017 at 5:24 | comment | added | SmallChess | @Moony Maybe she meant building logistic regression for many models or data set? | |
| Jan 19, 2017 at 5:05 | comment | added | Moony | Not really. I am running a logistic regression assessing timing of solid food intro (categorical months) with risk of incident diabetes(yes/no)- (crude unadjusted) then adding in covariates. Not sure why my colleague told me I would need to worry about multiple comparisons in this case? | |
| Jan 19, 2017 at 4:51 | comment | added | SmallChess | @Moony Multiple tests, not covariates. Correct. Is that what you are doing? | |
| Jan 19, 2017 at 4:44 | comment | added | Moony | I did read through the article, but I guess I might be misunderstanding because it looks to me like it applies more to having multiple outcomes, rather than having a lot of covariates? | |
| Jan 19, 2017 at 4:38 | comment | added | Michael R. Chernick | This is good. You left out the resampling approaches though. | |
| Jan 19, 2017 at 4:34 | history | answered | SmallChess | CC BY-SA 3.0 |