Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

Default priors bayesian GLMM

edited December 2022 in JASP & BayesFactor

Dear all,


I was wondering what the default prior specification is in JASP for the bayesian GLM with poisson family and log link since I can't seem to specify those myself. Are you also using the rstanarm priors that are autoscaled for the GLMM's?


Kind regards


Marijn

Comments

  • Hi Marijn,

    I'll ask our expert. This seems to be information that ought to be reported in the help file...

    Cheers,

    E.J.

  • Hi Marijn,

    Yes, the Bayesian (Generalized) Mixed Models analysis is using the autoscaled priors from rstanarm (as the analysis provides estimation only and no Bayes factors). We should've described this better in the documentation -- we are planing to make the prior specification accessible to the users and we will update the documentation accordingly.

    Cheers,

    Frantisek

Sign In or Register to comment.