Why does glmer have problems fitting gamma model?
This is because you scaled the response variable to be centred around zero and the gamma model is only for positive values. The other issues, that were fleshed out in the comments/chat is that glmer was having problems converging due to the way it approximates the integrals over the random effects in the definition of the marginal likelihood.
How to fit a gamma distribution in GLMM?
I am trying to fit a Gamma distribution (I ran the model with lmer but the residuals are not normally distributed) in a GLMM. My response variable is a list of numeric values (biomass), my fixed effects are two factors (with 2 levels each) and a date, my random effects are all factors.
Which is harder to Interpet gamma or linear model?
However this results in a model that is much harder to interpet than the linear model, while providing almost the same inferences and answers to the research questions. Having said all that, the root issue in the question is that the gamma model fitted clearly has very poor fit to the data.
Which is the best fit for a GLMM?
For a linear mixed-effects model (LMM), as fit by lmer, this integral can be evaluated exactly. For a GLMM the integral must be approximated. The most reliable approximation for GLMMs is adaptive Gauss-Hermite quadrature, at present implemented only for models with a single scalar random effect.
Are there different ways to parameterize the gamma distribution?
There are multiple ways to parameterize the Gamma distribution, so it’s important to pay attention when moving between languages and functions. As with many optimization exercises, we can force a term (here shape) to be positive by fitting in log-space.
What are the details of the glmer data set?
Data set details: Zeros are “real” (volume) Data set is heavily left skewed (even when zeros are excluded) Response is continuous (volume) Can anyone recommend a distribution family and link that I can use for glmer?