Bayesians: does anyone (@solomonkurz.bsky.social maybe?) have a good resource for how to set priors for beginners? Such as what kind of information in past papers you can use to inform your choices?

I found this but any resources welcome! svmiller.com/blog/2021/02/thinking-about-your-priors-bayesian-analysis/

March 26, 2026 - 10:09 UTC

10
1
36
Here’s a really good paper on the topic my grad advisor is on:

www.tandfonline.com/doi/abs/10.1080/15427609.2017.1370966
1
0
3
This is perfect, how has this not popped up in my searches?!
1
0
2
Going to add this one I came across recently: osf.io/preprints/psyarxiv/q7byw_v1
1
0
1
The author replied with this earlier, but thank you!
0
0
0
since there are no hard rules, I think the papers on Bayesian workflow/prior predictive checks are the most helpful imo. for cog sci, this paper is nice:

psycnet.apa.org/doi/10.1037/met0000275

see also this great (free) textbook:
bruno.nicenboim.me/bayescogsci/ch-compbda.html href="/search?t=posts&q=sec-priorpred">#sec-priorpred
1
0
7
Thanks! I hadn't seen the Schad one before.
0
0
0
We did a workshop covering that (4ccoxau.github.io/PriorsWorkshop/) and I have a lecture zooming out on priors and their impact www.youtube.com/watch?v=kMSrRd4f2As
1
0
8
That looks great, and you've just introduced me to `sample_prior = "only"` !
1
0
2
Conjugate priors can be a helpful starting point to build from
0
0
1
I'd say using past papers is advanced stuff. For beginners I'd use range/plausibility (weakly informative), e.g. outcome is blood pressure, you have lm, so e.g. N(150, 30) for intercept and N(0, 15) for between group difference captures the plausible range well enough (dtto for log/logit link).
1
0
1
That's really good advice, I like that.
1
0
1
The answer is always N(0,1)! 😉
0
0
1
Depending on which models you plan on using, we just shared this preprint: osf.io/preprints/psyarxiv/q7byw_v1
1
0
2
It's for students, so the more general the better as it's an introduction Bayesian modelling. This looks great though for a range of sources and models.
0
0
2
I like this chapter from Bayes Rules!

www.bayesrulesbook.com/chapter-4 href="/search?t=posts&q=ch4-priors">#ch4-priors
0
0
5
github.com/stan-dev/stan/wiki/prior-choice-recommendations
0
0
10