1

I am trying to understand how to specify (implement) an informed prior with brms in R. For example, a simple linear mixed effect model like:

DV ~ F1 * F2 + (1|Participant)

F1 and F2 are both two-level factors. I am only interested to test if one level of, say, F2 is significantly higher than another, either as the main effect only, or in interaction (e.g., a significantly stronger/steeper difference). In effect, this would be a case of one-sided testing, which should be more "meaningful" in the Bayesian framework. For the rest of the model, I do not have any specific prior expectations.

How should I define my informative prior with brms?

striatum
  • 1,428
  • 3
  • 14
  • 31
  • 3
    Doing a one-sided test is a perfectly reasonable request, but I don't think you would do it by setting an informed prior. (An informed prior that excluded negative values would say that you **assumed** the effect would be positive, not that you were interested in testing the probability that it was positive, or larger than a specified positive threshold (i.e. "positive and large"). – Ben Bolker Mar 16 '22 at 20:35

0 Answers0