Timeline for A repeated-measures LMM... Am I understanding this correctly?
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 27, 2020 at 6:19 | vote | accept | Jay | ||
| May 31, 2013 at 17:59 | comment | added | usεr11852 |
Yes, seems right for a first model. Just to clarify: lme(responsevar ~ -1 + treatment*sex ... would be if you did not wish an intercept to be included. I tried to use similar notation as in the book you mentioned, Sect. 2.2.2 and 2.3.2. are the ones that I found most relevant.
|
|
| May 31, 2013 at 17:30 | comment | added | Jay | Thanks, I realize now that I mistakenly labeled my response variable as a fixed effect. Whoops! I understand that the residuals need to be normally distributed, but I find myself a bit lost by your other explanations and comments (sorry! my brain just doesn't seem to work well with this stuff.) But essentially, my model in R would be something like: result <- lme(responsevar ~ treatment*sex, random = ~1 | SubjectID) Does that sound right? Thanks again! | |
| May 31, 2013 at 16:55 | history | edited | usεr11852 | CC BY-SA 3.0 |
Changed $\gamma$ to $u$ to be closer to the notation used in the book mentioned by the OP.
|
| May 31, 2013 at 16:50 | history | answered | usεr11852 | CC BY-SA 3.0 |