[R-sig-ME] Negative binomial glmm estimates backtransformation and 95% CI

Matteo Sebastianelli m@tteo@@b@ @end|ng |rom gm@||@com
Mon Jun 1 12:37:54 CEST 2020

Hi everyone,

I've been running a nbinom1 model in glmmTMB and need to show real values
for the estimates and 95% CI. Are estimates form nbinom1 glmmTMB
log-values? If so, is the transformation done by:

emmeans(model, "predictor", type="response")

Alternatively, calculating confidence intervals with "confint" and then
exponentiate them could be a possibility?



	[[alternative HTML version deleted]]

More information about the R-sig-mixed-models mailing list