Default priors

Depends on the model

For most models:

  • intercepts are treated differently
  • regression coefficients have mean zero with some specific variance
  • scale parameters (e.g. residual variance) will have appropriate priors

Basically, rstanarm is going to be okay for you to use defaults

If you want to change, there are plenty of resources about priors:

?prior_summary

?priors

http://mc-stan.org/rstanarm/articles/priors.html

https://github.com/stan-dev/stan/wiki/Prior-Choice-Recommendations