Predict with brms fit object

Interfaces brms
1

Please also provide the following information in addition to your question:

  • Operating System: MacOS High Sierra
  • brms Version: 2.7.0

I’m running a multilevel model in Stan with one binary predictor and 1 factor-type random intercept and a truncated outcome variable - i.e y | trunc(lb=10, ub = 50) ~ x1 + (1|x2). The model runs fine.

However, I wanted to do some posterior checks. The data looks like this:

image
image.png774×610 7.82 KB

At first I ran predict() on the brms fit object and this was the distribution of the data
brms_predict
brms_predict.png522×534 6.32 KB

I don’t think this is an issue with model specification because the pp_check() on the same object looks like this
Brms_ppcheck
Brms_ppcheck.png522×534 31.1 KB

I thought predict worked on a brms object in same way posterior_predict works on the rstanarm fit objects - have I misunderstood?

Thanks! :)

2

You may use posterior_predict also with brms or predict(..., summary = FALSE).

3

Great thanks @paul.buerkner - I knew I was doing something silly!