Is prior predictive check possible ? #229
Unanswered
osupplisson
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi,
I finally found out some georeferenced data and decided to give inlabru an in-depth try: truly amazing work !
To ensure my priors are not too dumb, I would like to perform a 'prior predictive' check (https://mc-stan.org/docs/stan-users-guide/posterior-predictive-checks.html). Is there any hidden function for that in the package?
I tried with the
bru_run = FALSE
option + thepredict()
function but it did not work. I haven't found anything else to try in bru (nor in INLA). Am I missing a 'hidden' function for sampling from the initial configuration (i.e., predict from the prior model)?Is that even theoretically possible with the R-INLA approach or is it only a matter of implementation?
Best,
Olivier
Beta Was this translation helpful? Give feedback.
All reactions