Timeline for Regression with 3 measurement points (in R)
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 30, 2013 at 16:03 | comment | added | Torvon | In addition, would you know of one test giving one p-value that can test the assumption that symptoms differ from each other in variance explained? I have 14 values (between 0.1% and 19.8%) of explained variance, but don't have them per subject (so I cannot use tests like Hotelling's T). | |
| Jan 30, 2013 at 11:22 | vote | accept | Torvon | ||
| Jan 30, 2013 at 11:21 | comment | added | Torvon | Thank you. (1) I wouldn't know how to write an R script for that. Option (2) would also take into account all 3 times, in a less proper fashion than option (1). I'd be happy about recommendations when it comes to alternative methods of R^2 decomposition - my question is whether 14 symptoms of the disorder X are differentially associated with impairment. Obtaining individual R^2 contribution sounded like the way to go (if one has 20% variance explained and the other 1% and the CIs are 2%, that's pretty substantial). | |
| Jan 30, 2013 at 11:17 | history | bounty ended | Torvon | ||
| Jan 27, 2013 at 19:00 | history | edited | Peter Ellis | CC BY-SA 3.0 |
added 30 characters in body
|
| Jan 27, 2013 at 18:53 | history | answered | Peter Ellis | CC BY-SA 3.0 |