From 0df7dc45a946681146dcc0c569c3d091a215bdaf Mon Sep 17 00:00:00 2001 From: jgabry Date: Wed, 10 Jun 2026 16:40:47 +0200 Subject: [PATCH] Mention new loo_compare columns in vignette --- vignettes/loo2-example.Rmd | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/vignettes/loo2-example.Rmd b/vignettes/loo2-example.Rmd index cf32fb8f..2a5037a7 100644 --- a/vignettes/loo2-example.Rmd +++ b/vignettes/loo2-example.Rmd @@ -272,11 +272,15 @@ loo_compare(loo1, loo2) The difference in ELPD is much larger than several times the estimated standard error of the difference again indicating that the negative-binomial model is -xpected to have better predictive performance than the Poisson model. However, +expected to have better predictive performance than the Poisson model. However, according to the LOO-PIT checks there is still some misspecification, and a reasonable guess is that a hurdle or zero-inflated model would be an improvement (we leave that for another case study). +The `p_worse`, `diag_diff`, and `diag_elpd` columns are new as of version 2.10. +For details on interpreting them see ?`loo-glossary` and the case study +[Uncertainty in Bayesian LOO-CV Model Comparison](https://users.aalto.fi/~ave/casestudies/LOO_uncertainty/loo_uncertainty.html). +