Timeline for R: How to fit a linear mixed model with a custom covariance structure for two random intercepts
Current License: CC BY-SA 4.0
2 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 10 at 11:22 | comment | added | Giora Simchoni | Thank you for your reply. If instead of just $b_{1j}+b_{2j}$ the model would have been non-linear (NLMM) like $f_1(b_{1j})+ f_2(b_{2j})$, where $f_1, f_2$ are some functions, it would have been identifiable, right? Any other way which comes to mind? | |
| Dec 9 at 22:02 | history | answered | statmerkur | CC BY-SA 4.0 |