okay #rstats #rstan #stan hivemind:

do you have any examples of Stan models (incl #brms) running in production, especially attached to Shiny apps where responsiveness/compute time is pretty important (and interfacing with non-quant people)?

What tricks do you use?

Please send blogs, packages, repos, anecdotes! :)

Please do not send: suggestions that I use an empirical Bayes/frequentist framework. I know how to do that :)

📰 Happy to announce the publication in @PeerCommunityJournal of our analysis of #VaginalMicrobiota dynamics.

🦠 This work provides unprecedented insights into the occurrence of #dysbioses.

Thanks to the #women participating in the #PAPCLEAR study at CHU Montpellier (France) 🙏

The model (all in #RStan) was built by @tsukushi_kamiya and the #16S sequencing of over 2.000 samples was done in Jacques Ravel's lab. This work stems from the @ERC_Research #EVOLPROOF project.

https://peercommunityjournal.org/articles/10.24072/pcjournal.527/

Heureux d'annoncer la publication dans @PeerCommunityJournal de notre analyse de la dynamique du #MicrobioteVaginal.

Ces travaux sont issus de l'étude #PAPCLEAR promue par le #CHU_Montpellier et financée par l' @ERC_Research, qui représente l'un des suivis les plus longs à ce jour.

Le magnifique modèle statistique (tout en #RStan) a été réalisé par @tsukushi_kamiya et le séquençage #16S de plus de 2000 échantillons réalisé dans le laboratoire de Jacques Ravel.

https://peercommunityjournal.org/articles/10.24072/pcjournal.527/

#statstab #181 Comparing Bayesian Approaches by @jebyrnes

Thoughts: Compares running models in #rethinking, #stan, #brms, #inla, and #glmmTMB via #TMBstan. It's nice to have options.

#bayes #bayesian #rstan #r #stats #rstats

https://biol609.github.io/lab/alt_to_rethinking.html

#statstab #169 {priorsense} prior diagnostics and sensitivity analysis

Thoughts: Bayesian modelling requires more scrutiny of how one's choices impact outcomes. This packages has handy functions + plots.

#brms #bayesian #rstan #priors
#mcmc #diagnostics

https://n-kall.github.io/priorsense/

Обзор библиотеки Stan в R

Приветствую! Stan - это библиотека на C++, предназначенная для байесовского моделирования и вывода. Она использует сэмплер NUTS, чтобы создавать апостериорные симуляции модели, основываясь на заданных пользователем моделях и данных. Так же Stan может использовать алгоритм оптимизации LBFGS для максимизации целевой функции, к примеру как логарифмическое правдоподобие . Для облегчения работы с Stan из языка программирования R доступен пакет rstan , который предоставляет интерфейс R для Stan. Сегодня мы и рассмотрим этот пакет.

https://habr.com/ru/companies/otus/articles/794196/

#rstan #r #аналитика #машинное_обучение #c++

If I am a toddler in my #statistics development, then I could be said to be a baby in #Rstats, but I haven't even had my umbilical cord cut in #bayes. I installed #rstan this morning, then ran the schools.stan example.

My laptop didn't seem to do anything so I assumed that I had installed incorrectly and so went to make a cup of tea.

Imagine my surprise when I returned to a screen full of all sorts of stuff.

I am the noobiest of noobs. A lot to learn!

You've heard about Stan and want to learn some more? Maybe you're about to step into the Bayesian paradigm and don't know where to start 🤔

In this week's blog post, we'll take a look what you can do with #Stan and draw comparisons to #JAGS!

https://jumpingrivers.com/blog/why-stan/

#DataScience #RStats #Python #RStan #PyStan #Bayesian