Aki Vehtari

As the above minimal example doesn't currently produce zombies for me, here is a new minimal example ``` library(brms) options(brms.backend = "cmdstanr", mc.cores = 2) m0

I think Andrew's comments on this are relevant. See a recent blog post https://statmodeling.stat.columbia.edu/2025/05/11/plotting-truth-vs-predicted-value/ and section 11.3 of Regression and Other Stories: “A confusing choice: plot residuals vs. predicted values,...

Mean in `ppc_stat()` is completely different thing from mean in `ppc_error_scatter_avg()`, and the reason which makes mean in `ppc_stat()` a bad choice, does not make it make in `ppc_error_scatter_avg()`.

> @avehtari I think that chapter in RAOS argues for doing what @kruschke suggests, right? Yes. > In those figures is the residual computed as y - stat(y_rep) or stat(y...

You might also consider using `posterior` package, https://github.com/stan-dev/posterior, for convergence diagnostics. Its repo is under github.com/stan-dev, but it has been designed to be Stan agnostic, and can be used without...

I'm a co-author of the `posterior` package. My contributions are mainly in the diagnostics (which are also better than in coda, see https://doi.org/10.1214/20-BA1221 and https://avehtari.github.io/rhat_ess/ess_comparison.html) We're happy to help if...

> @avehtari @jgabry do we also want to implement the MCSE described in section 4.4 ? has this already been done in R? which package and where? > Yeah bulk...

I agree that that FIXME comment should be removed

Thanks for reporting this. Unfortunately, we can't solve this in `loo` package side. We can improve the message in `loo` in case of `generated quantities` block causing errors, but that...

1. I think it fits here. It's still generic so that there is no dependency on where the draws come from, and if pit function would support draws objects it...