Add missing citation (#191)

Author: aloctavodia

Chapters/Prior_posterior_predictive_checks.qmd

@@ -718,4 +718,5 @@ dt_cens = azp.load_arviz_data('censored_cats')
 azp.plot_ppc_censored(dt_cens)
 ```
 
-To keep predictions within a meaningful range ArviZ limits the predictive data relative to the maximum observed duration, preventing the survival curves from extending too far beyond the observed data. You can change this with the argument `extrapolation_factor`, which defaults to 1.2, meaning up to a 20% larger that the maximum observation.
+To keep predictions within a meaningful range ArviZ limits the predictive data relative to the maximum observed duration, preventing the survival curves from extending too far beyond the observed data. You can change this with the argument `extrapolation_factor`, which defaults to 1.2, meaning up to a 20% larger that the maximum observation. For more recommendations on how to perform predictive assessment for survival model you can read [@suorsa_2026].
+

references.bib

@@ -888,3 +888,13 @@ @misc{soch_2024
 	doi={10.5281/ZENODO.4305949},
 	url={https://statproofbook.github.io/}
 }
+
+@misc{suorsa_2026,
+      title={Predictive Assessment and Comparison of Bayesian Survival Models for Cancer Recurrence}, 
+      author={Saku Suorsa and Aki Vehtari},
+      year={2026},
+      eprint={2601.01662},
+      archivePrefix={arXiv},
+      primaryClass={stat.ME},
+      url={https://arxiv.org/abs/2601.01662}, 
+}