Timeline for R: lme4 vs. glmmTMB for binomial GLMM
Current License: CC BY-SA 4.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 24, 2021 at 18:00 | history | tweeted | twitter.com/StackStats/status/1386017113891123202 | ||
| Apr 30, 2019 at 13:49 | comment | added | amoeba |
It's still linear. You can start with glm and include sin(month/12 * pi) and cos(month/12 * pi) as fixed effects. More fancily one would use cyclic splines and gam. But I am not an expert in any of that.
|
|
| Apr 30, 2019 at 11:33 | comment | added | Joris | Thanks. I'm not very familiar with nonlinear regression - any suggestions for packages/resources to get started on this? | |
| Apr 30, 2019 at 11:18 | comment | added | amoeba |
It looks like glmmTMB might not be handling weights correctly. However, I'm not sure that modeling month as a random effect is very meaningful. I think it should rather be fixed, ideally treated as a circular variable using polynomial terms or splines.
|
|
| Apr 30, 2019 at 10:44 | history | asked | Joris | CC BY-SA 4.0 |