In the Stan wiki there is currently this quote here. [image] Prior Choice Recommendations Stan development repository. The master branch contains the current release. The develop branch contains the latest stable development. See the Developer Process Wiki for details. - stan-dev/stan …

I don’t remember why Aki liked 3 in particular for the degrees of freedom, but the appeal of the t-distribution for priors is that the degrees of freedom parameter functions like a dial you can turn to make the distribution more Cauchy-like (df -> 1) or more Gaussian-like (df -> Inf). A small degree…

Having degrees of freedom greater than 1 ensures a finite variance and having it greater than 2 ensures a finite mean, as well. So between 2 and 3 is when your tails are only incredibly heavy but not obscenely heavy, which is what I would guess was Aki’s reasoning. People argue about values rangin…

Thanks for the responses! I definitely learned stuff

[image] betanalpha: the benefits of heavier tails don’t outweigh their complications. Could you expand a little bit on your reasoning? From your case study , I took away that the normal could be too constraining, compared to heavier tailed priors. At least that is what I conclude from the fail…

[image] betanalpha: Having degrees of freedom greater than 1 ensures a finite variance and having it greater than 2 ensures a finite mean, as well. So between 2 and 3 is when your tails are only incredibly heavy but not obscenely heavy, which is what I would guess was Aki’s reasoning. Yes. Th…

The weakly-informative priors that we typically talk about enforce “containment” of the posterior. The shape of the prior density impacts just how strong this containment is. Lighter tails, like Gaussians, offer stronger containment. This prevents the posterior from stretching to extreme regions …

Why StudentT(3,0,1) for prior?

General

anon79882417 March 16, 2019, 5:15pm 2

@avehtari

N best tips & tricks (or the go-to checklist) for new Stan model builders?

Topic		Replies	Views
Estimating student-t degrees of freedom? Modeling	2	2070	January 2, 2018
PSA: Marginally-Student-t multivariate (df-varying-by-dimension) Modeling specification	8	656	June 17, 2022
Non-centered student_t - matt's trick Modeling techniques	2	1516	July 2, 2018
Student T distribution: A question on scale General	3	3642	July 29, 2020
Large estimated standard deviation in multilevel model despite weakly informative prior Modeling	7	760	March 24, 2019

Why StudentT(3,0,1) for prior?

Related topics