Priors documentation

[strengejacke]

Currently, the docs for the coef argument in priors says:

The exception is for fixed-effect parameter vectors, where the intercept (if present) is not included; the prior on the intercept must be set explicitly.

We discussed whether or not the intercept should be excluded by default for other components as well - not sure how you decided, but if so, the documentation is not up to date, right?

[bbolker]

The documentation is correct, but could be clarified/expanded; "fixed-effect parameter vectors" means the fixed effects for all three model components (conditional, zero-inflated, and dispersion models), as opposed to the random-effect parameter vectors.

Does that make sense?

[strengejacke]

Yes, makes sense. It's probably my lack of statistical education that I sometimes struggle with common statistical terms - but also to avoid possible confusion with econometrics and their fixed effects, this clarification might be helpful.