# Poor estimates from pymc examples on marginal and latent GP

**URL:** <https://discourse.pymc.io/t/poor-estimates-from-pymc-examples-on-marginal-and-latent-gp/12204>\
**Category:** v5\
**Tags:** gaussian\_process, modeling\
**Created:** [May 25, 2023, 2:42pm UTC](https://discourse.pymc.io/t/poor-estimates-from-pymc-examples-on-marginal-and-latent-gp/12204 "2023-05-25T14:42:25Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![giorgp](https://avatars.discourse-cdn.com/v4/letter/g/b19c9b/32.png) [@giorgp](https://discourse.pymc.io/u/giorgp)\
**Post date:** [May 26, 2023, 7:19pm UTC](https://discourse.pymc.io/t/poor-estimates-from-pymc-examples-on-marginal-and-latent-gp/12204/3 "2023-05-26T19:19:13Z")

</div>

Many thanks for your insightful response, your explanation makes sense. As mentioned above, I went through these examples in order to investigate [some modelling I had done in Stan](https://discourse.pymc.io/t/computer-model-calibration-using-a-gaussian-process-from-stan-to-pymc/12180).

For reference, earlier today I reproduced the marginal GP example discussed in this post in Stan and got similar results as in pymc.

---

_[View the full topic](https://discourse.pymc.io/t/poor-estimates-from-pymc-examples-on-marginal-and-latent-gp/12204)._