# Using narrower priors for SBC

**URL:** <https://discourse.mc-stan.org/t/using-narrower-priors-for-sbc/21709>\
**Category:** General\
**Tags:** simulation-based-calibration, prior-choice\
**Created:** [April 4, 2021, 1:33pm UTC](https://discourse.mc-stan.org/t/using-narrower-priors-for-sbc/21709 "2021-04-04T13:33:49Z")\
**Posts on this page:** 1\
**Showing post:** 2

<div class="post-metadata">

**Author:** ![betanalpha](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/betanalpha/32/4_2.png) [@betanalpha](https://discourse.mc-stan.org/u/betanalpha)\
**Post date:** [April 5, 2021, 7:04pm UTC](https://discourse.mc-stan.org/t/using-narrower-priors-for-sbc/21709/2 "2021-04-05T19:04:24Z")

</div>

> [@SBC for arma model](https://discourse.mc-stan.org/t/sbc-for-arma-model/21362/9):
>
> I had assumed since the benchmark models have been extensively tested that there shouldn’t be any issues, but that only applies for the given datasets.

Each statistical computation benchmark is a specific posterior distribution with validated posterior expectation values _not_ a generic model.

SBC tests the consistency of a computational method of an entire ensemble of posterior distributions generated from prior predictive data which (usually, although not always) requires that the computational method fits _all_ of those posteriors well. If the priors are wide then more and more posterior behaviors will have to be accommodated which may frustrate the given computational method even if some of the posteriors distributions fit fine.

---

_[View the full topic](https://discourse.mc-stan.org/t/using-narrower-priors-for-sbc/21709)._