Mixed effects models - clusters with one observation

Asked Jan 05 '23 at 21:31

Active Jan 07 '23 at 15:11

Viewed 64 times

I'm quite new with GLMM, and I'm fitting a linear mixed model in R since I'm dealing with longitudinal data. I want to treat subjects id as random effect.

I have 290 clusters made up by subjects observed from one to 14 times, and a relative large number (45) are observed only once, should I drop them ? I'd not since they are many. Is that still okay having many subjects observed once?

edited Jan 06 '23 at 09:30

asked Jan 05 '23 at 21:31

Maximilian

Welcome to CV Maximilian. Unfortunately, I am recommending closure for your question: The way CV works, you have the basis of three separate questions here (1 from each paragraph). I would recommend editing your question ("edit" link at lower left) to remove the second two questions, and ask those as two separate new questions. – Alexis Jan 06 '23 at 07:40
1

Question amended. Could you pls unlock the question? – Maximilian Jan 06 '23 at 09:30
Done! :D Also look around the site, I know there have been similar questions. – Alexis Jan 06 '23 at 18:07
1

Minimum sample size per cluster in a random effect model – dipetkov Jan 07 '23 at 15:21
Thanks so much, I think that the question already answered and just flagged in the last comment is quite exhaustive. – Maximilian Jan 08 '23 at 13:37

Mixed effects models - clusters with one observation

0 Answers0