# Sum to zero for BYM2 unstructured effect

**URL:** <https://discourse.mc-stan.org/t/sum-to-zero-for-bym2-unstructured-effect/40277>\
**Category:** Modeling\
**Tags:** specification\
**Created:** [August 24, 2025, 8:54am UTC](https://discourse.mc-stan.org/t/sum-to-zero-for-bym2-unstructured-effect/40277 "2025-08-24T08:54:02Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![RobertoCerina](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.mc-stan.org/robertocerina/32/20357_2.png) [@RobertoCerina](https://discourse.mc-stan.org/u/RobertoCerina)\
**Post date:** [August 24, 2025, 5:44pm UTC](https://discourse.mc-stan.org/t/sum-to-zero-for-bym2-unstructured-effect/40277/3 "2025-08-24T17:44:57Z")

</div>

Thanks @mitzimorris – my understanding is that \theta \times \sigma is basically a random effect specified via non-centred parametrisation, and as such it’s “soft” centred (correct me if I’m wrong). Based on my reading of [this](https://discourse.mc-stan.org/t/proper-use-of-sum-to-zero-vector-in-nested-multilevel-models/39536/18) thread, I had understood that it was preferable for posterior-sampling efficiency gains to specify traditional random intercepts with the sum\_to\_zero formulation where possible. Did I get that wrong ? Thanks again.

---

_[View the full topic](https://discourse.mc-stan.org/t/sum-to-zero-for-bym2-unstructured-effect/40277)._