Weakly informative, default priors for linear regression models?

Dear all, what weakly informative (or non-informative) default priors are recommended for linear regression models with normal outcome? I found the pubs by Gelman et al. (2008, 2009), but I am not sure this approach is still recommended. What default priors for linear regression are available in R/brms and Stan? Advice welcome!

Kr, Shiny

Refs:

We have a convenient summary in the form of a web page:

There are more recommendations scattered through the User’s Guide. You can ask about specifics here on the forums.

We do not recommend the Cauchy priors that Aleks Jakulin et al. used in those cited papers. We do recommend standardizing predictors.

2 Likes