Singular Fit Error in Generalized Mixed Model

General help and assistance with jamovi. Bug reports can be made at our issues page: https://github.com/jamovi/jamovi/issues . (If you're unsure feel free to discuss it here)

by akacar » Wed Nov 11, 2020 8:54 pm

Hello,

I'm trying to conduct a generalized mixed model analysis (logistic). I do not encounter any error with jamovi. But when I run the same analysis with R, it results in singular fit error. Do you have any suggestions about why this discrepancy might emerge?

I attached the jamovi result. The code that I use in R is below.

glmer(stay ~ transition2*reward2 + (transition2*reward2 + 1|participantno), data=data_mel3, family="binomial")

Thank you,
Anıl Şafak Kaçar
Attachments
Capture.JPG
Capture.JPG (41.09 KiB) Viewed 97 times
akacar
 
Posts: 3
Joined: Mon Nov 26, 2018 10:11 am

by mcfanda@gmail.com » Wed Nov 11, 2020 9:21 pm

The latest version of glmer() warns you for "near" singular fit when using the default optimizer. GAMLj, when it finds possible singular fit, changes the optimizer to find a better solution. If you do not see a warning in GAMLj, it means it found an optimizer that produced an estimation with no singular fit.
let me know if that does not work for you
User avatar
mcfanda@gmail.com
 
Posts: 217
Joined: Thu Mar 23, 2017 9:24 pm

by akacar » Wed Nov 11, 2020 9:25 pm

Thank you very much for your quick response. Is there any way to see which optimizer is used by GAMLj?
akacar
 
Posts: 3
Joined: Mon Nov 26, 2018 10:11 am

by mcfanda@gmail.com » Wed Nov 11, 2020 9:41 pm

that's a good question. I'm afraid it does not say it in the output. We can add it as a footnote in the next version. We do try to keep the output clean, so we avoid adding information that is not essential.
User avatar
mcfanda@gmail.com
 
Posts: 217
Joined: Thu Mar 23, 2017 9:24 pm


Return to Help