Coeff plots for brms

I am playing with some options to plot posterior samples — what I think of as coefficient plots. I can get densities with ggridges::geom_density_ridges(), but I have found other resources such as brmstools::forest and coefplot. The problem with the last two is I get the error.

Error: No group-level effects detected. Call method ‘fixef’ to access population-level effects.

I have two questions:

  1. What is the reason for the above error?

  2. What are some other popular ways to plot coefficients?

The latter two are using the coef() method of brms which expects varying (“random”) effects to be present, because that’s what these functions are made for.

1 Like

What libraries would you use for models with only fixed effects?

The bayesplot package has quite a few options. See ?"MCMC-overview" after loading bayesplot. You can also access its functionality directly with brmsfit object via the stanplot method

1 Like