Poslovitch · @Poslovitch
98 followers · 795 posts · Server wikis.world

C'est fantastique l'info trafic en temps réel de à : maintenant que TwitiX () est un feu de poubelle, impossible de lire quoi que ce soit. Par contre je découvre qu'en 2021 a utilisé le réseau pour distribuer des oranges, qu'en 2022 un agresseur a été arrêté...

L'info trafic en temps réel vous disais-je ? Plutôt "Souvenirs de STAN" !

#Stan #nancy #twitter #saintnicolas

Last updated 1 year ago

Gabriel Weindel · @GWeindel
195 followers · 111 posts · Server fediscience.org

Finally a implementation for the full
Looking forward to get some chains running (hopefully for not too long)

link.springer.com/article/10.3

#driftdiffusionmodel #Stan

Last updated 1 year ago

Bruno Nicenboim · @bruno_nicenboim
428 followers · 272 posts · Server fediscience.org

I don't think I shared this: I wrote a short paper for the last ICCM/Mathpsych (mathpsych.org/presentation/998).

It's a computational model of self-paced reading (written in @mcmc_stan) in a framework of continuous flow of information.

@cognition @linguistics @psycholinguistics @cogsci

#ComputationalModeling #cogsci #rstats #Stan #bayesian #psycholinguistics

Last updated 1 year ago

Martin Modrák · @modrak_m
400 followers · 627 posts · Server fediscience.org

New version of cmdstanr exposes Stan functions and allows you to directly unconstrain parameters (super useful for seeing parameters as the sampler sees them)! A huge win!

mc-stan.org/cmdstanr/news/inde

#rstats #bayes #mcmc_stan #cmdstanr #Stan

Last updated 1 year ago

Martin Modrák · @modrak_m
386 followers · 573 posts · Server fediscience.org

Updated preprint on validating that your Bayesian model and/or sampling algorithm is working correctly.
arxiv.org/abs/2211.02383

Thread about the first version: fediscience.org/@modrak_m/1093

We've added a section where we use SBC to build an ordered-simplex data type for Stan. We've also done minor improvements based on reviewer feedback and thanks to F. Saad alerting us that we misrepresented their work previously (my fault).

Thx @avehtari and all other co-authors not on here.

#Stan #sbc #bayesian

Last updated 1 year ago

Tomáš Zdechovský · @TomasZdechovsky
73 followers · 6104 posts · Server respublicae.eu

Samozřejmě, že @danusenerudova je nejlepší volba na eurokomisařku za

Osobně to rozhodně podporuji! 👏👏👏

🐦🔗: n.respublicae.eu/TomasZdechovs

#Stan

Last updated 1 year ago

Piia Akka_69 · @Piia
266 followers · 682 posts · Server eliitin-some.fi

Multa kysyttiin tännään, että koska viimeksi olen käynyt erään tietyn pariskunnan luona kyläsä? En ole käyny koskaan heillä. Kysyjä oli olettanut, että me käyään toistemme luona kyläsä, ko tollaan toisen osapuolen kans saman työnantajan palaveluksesa ja asutaan samasa kerrostalosa.

Tota noi...En muutenkaan ole kyläilijätyyppiä; harvan päästän luokseni ja lähes yhtä harvan luona käyn kylässä. On muitaki tapoja tavata ystäviään/kavereita. Etenki ko suurimmaksi osaksi assuu kaukana

#Stan

Last updated 2 years ago

Afroféminas · @afrofeminas
24 followers · 99 posts · Server federated.press
Martin Modrák · @modrak_m
374 followers · 428 posts · Server fediscience.org

I am looking for
a) examples of tools that let you build statistical models more complex then just variations of a single model class (like most stat packages - brms, laavan, ...) but less complex than fully fledged probabilistic programming languages
b) Probabilistic programming languages that neatly support composing non-trivial submodels together

Does anyone have recs?
In both cases I am coming up almost empty handed...

#brms #ProbabilisticProgramming #ppl #Stan

Last updated 2 years ago

Maximilian Scholz · @scholzmx
76 followers · 82 posts · Server fediscience.org

After a small thing involving Bayesfactors, I am now convinced that they are the main reason for the opinion that Bayesian statistics is subjective mumbo jumbo and I can totally understand the opinion from that perspective 😂

#brms #bayes #Stan

Last updated 2 years ago

Maximilian Scholz · @scholzmx
75 followers · 80 posts · Server fediscience.org

We have experimental Support for the symlog-normal distribution in bayesfam. A continuous unbounded transformed normal distribution using the symlog link from arxiv.org/abs/2301.04104 .
Seems to be able to fit skewed- as well as pointy assymetric-laplace style data.
github.com/sims1253/bayesfam/p

@rstats

#Stan #brms

Last updated 2 years ago

Zane Billings · @wzbillings
58 followers · 100 posts · Server fediscience.org

Is there a way to print the contents of a file in and have the output use correct syntax highlighting??

#Quarto #Stan

Last updated 2 years ago

Maximilian Scholz · @scholzmx
75 followers · 78 posts · Server fediscience.org

I have a failing r-cmd-check action on windows in github.com/sims1253/bayesfam/a that cries about `Error in `sink(type = "output")`: invalid connection` and seemingly has something to do with rtools and/or tbb. Does anyone have experience with Gh actions for packages that rely on rstan? @rstats

#Stan

Last updated 2 years ago

Manfred Weber · @ManfredWeber
18 followers · 372 posts · Server respublicae.eu

Thank you for the talks @MarekVyborny, @kducsl, , @TOP09 colleagues. Congratulations to everyone who made the Czech Presidency a success story. War, migration, the US Inflation Reduction Act. We must give European answers for a safe and prosperous future for our people.

🐦🔗: n.respublicae.eu/ManfredWeber/

#Stan

Last updated 2 years ago

Stop That · @stopthatgirl7
285 followers · 3978 posts · Server famichiki.jp

So this lady on the train had decided her bags needed their own seat on this full train.

And then this other woman got on, came over to sit in what looked from a distance like an empty seat, then just stood there and STARED very pointedly at the bags until the other woman moved them onto her lap.

She sat down, and I stan. 🤣

#onthetrain #publictransport #Stan

Last updated 2 years ago

Ilja Kocken · @japhir
143 followers · 63 posts · Server fediscience.org

@rlmcelreath I attended this course 2 years ago and it has changed the way I view statistical inference, how I think about doing analyses, as well as how to teach complex mathematical concepts in an understandable way. 10/10, would recommend! Adding some tags so that it may reach more interested people

#Stan #r #stats #statistics #teaching #bayesian #bayes #CausalInference

Last updated 2 years ago

Looking for a last minute holiday gift? Or just something with which to treat yourself? Give the gift of statistical education, events.eventzilla.net/e/princi!

Or find something they'd actually appreciate and then sign them up.

#Stan #bayes #modeling #statistics

Last updated 2 years ago

Real Quack · @ducker
43 followers · 738 posts · Server union.place

@lolennui

I Teddy Roosevelt and I am not apologetic about it.
(of course, he left the party)

#Stan #republican

Last updated 2 years ago

Jesse Wolfhagen · @uncertainarch
5 followers · 1 posts · Server archaeo.social

A belated

I'm Jesse Wolfhagen. I call myself a "Bayesian Zooarchaeologist" because I use Bayesian statistics to model different aspects of animal data from archaeological sites: from probabilistic taxonomic IDs to the age/sex composition of an assemblage to season of birth.

I am Interested in critically examining the way we view domestication processes and tell narratives about the past.

I also (try to) build rigorous statistical models for archaeology using and

#introduction #rstats #Stan

Last updated 2 years ago

Dieter Lukas · @DieterLukas
401 followers · 111 posts · Server fediscience.org

To analyse the data, we implemented Bayesian phylogenetic models in . We developed them based on helpful advice from here bookdown.org/content/3890/miss and the rethinking materials from @rlmcelreath github.com/rmcelreath/stat_ret

The approach gave the same results as metafor (phew), but the models allowed more flexibility, for example to better understand how effects differ across the phylogeny and to look at ->

#Stan #metaanalysis

Last updated 2 years ago