We have a new paper out in Policy Studies Journal.

https://onlinelibrary.wiley.com/doi/10.1111/psj.12515

In a nutshell, it's about governance networks in ten Swiss wetlands and looking at which actors within these networks are satisfied with their inclusion.
--------
Meta 1:
My #policy studies community seems largely stuck on X 🤮, so two technical things that might be interesting beyond the content:
- There's DAGs and #Bayesian models and #brms and #rstats <3.
- Reproducibility via a Docker container, a first for me.

stuck on a tricky issue with multinomial logit models with brms: https://discourse.mc-stan.org/t/use-categorical-multinomial-family-with-binary-ish-data-based-on-separate-column-in-brms/32383 #rstats #brms #bayesian

wrote some raw stan code for the first time in awhile (usually just #brms it these days), please clap

#rstats

#rstats (#brms) question: Say you measured participants' reaction times in two conditions A and B (within-person); how would you estimate different between-person variances for conditions A and B in a multilevel model?

Me 24 hours before my scheduled conference presentation:

#stan #brms #bayesian

Me 24 hours before my scheduled conference presentation:

#stan #brms #bayesian

Is it just me who doesn't like to be repeatedly told to: "Adjust [my] expectations accordingly"? Feels like an oddly intrusive thing for my #Rstats console to be telling me to do, especially when using an imperative and an exclamation mark! #Bayesian #brms

ahhh this new #bayesian ordered beta regression model family (package: https://github.com/saudiwin/ordbetareg_pack; paper: https://doi.org/10.1017/pan.2022.20) is so so neat! I have an outcome bounded at 1 and 32, and the model successfully predicts discrete 1s and 32s, as well as a continuous range in between! #rstats #brms #statsodon

Here is material for an intermediate course on #Bayesian #regression modeling using #brms : https://michael-franke.github.io/Bayesian-Regression/

Covers:

- prior / posterior model checking
- generalized linear & non-linear models
- mixture and distributional models
- MCMC methods (some Stan)
- model comparison
- causal inference

Stan developer Mitzi Morris gave R-Ladies NYC talk on how you can quickly build robust models for data analysis and prediction using `brms` and Stan, quick review of multi-level regression, and how to fit, visualize, and test the goodness of the model and resulting estimates. Video is available at https://www.youtube.com/watch?v=A1NWoKQhgJE

#Rladies #brms #Bayes #Stan

#RStats issues I'm struggling with that seem impossible to Google: Building a {brms} model within the {tidymodels} framework using {bayesian}.

The formula is inherently too complex (including splines and random effects) for the typical tidymodels workflow that involves recipes &c., so it must be added in at a later step. Two things:

1. Complex {brms} multivariate formulas seem to not be possible using {tidymodels}. E.g., literally multivariate or including phi after my formula via brms::bf(). It simply errors. :( This may just need some tweaking of {bayesian}'s scripts or waiting for an update since it's still fairly young.

2. Using {mgcv} random effect syntax like s(cat1, cat2, bs = "re") seems to not pick up as random effects in the model...I think? And I have never figured out if this is creating hierarchical random effects or not -- or if multilevel random effects just aren't possible in this syntax(?).

3. Using {lme4} random effect like (1 | cat1 / cat2) to ensure the hierarchy is preserved *does* retain random effects I can pull out of the model later using `ranef`, but for some absurd reason I cannot run this model through cross-validation or a myriad of other steps later because it seems to force-create a complex web of interacting factor levels that don't exist. E.g., if my random effects are '(1 | realm / biome)', this eventually fails because it'll look for tundra biome types in Africa for some absurd reason.*

Noticed this while trying to solve *separate* issues within broom.mixed:::tidy.brmsfit() -- that it seems to delete the names of all the fixed effects and return them as 'NULL' character strings (???), and its reliance on 'ranef' means it doesn't find the random effects using {mgcv} syntax.

That's my rambling mess of an essay for the day. Not sure how many of these are real issues or me simply not understanding how these packages differ or wot.

#brms #mgcv #tidymodels

* Almost wondering if this might even be a separate {tidymodels} issue right now. Every recipe no matter what seems to factor every single character column regardless of how the recipe is built. Hmmmm.

Big shoutout to our instructor @SolomonKurz & this interactive group of attendees for a fantastic course on Bayesian statistics using brms and the tidyverse🙌📈

#BayesianStatistics #brms #Tidyverse #Rstats

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...

#stan #ppl #ProbabilisticProgramming #brms

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 😂
#stan #bayes #brms

We have experimental Support for the symlog-normal distribution in bayesfam. A continuous unbounded transformed normal distribution using the symlog link from https://arxiv.org/abs/2301.04104 .
Seems to be able to fit skewed- as well as pointy assymetric-laplace style data.
https://github.com/sims1253/bayesfam/pull/14

@rstats #brms #stan

Verification of business rules programs
(2012) : Berstel-Da Silva, Bruno
url: https://nbn-resolving.org/urn:nbn:de:bsz:25-opus-87991
#production_system #rule_system #BRMS #formal_methods #verification #rule #my_bibtex

🧐All our last courses in February are #soldout. But there are still a few seats available for some of our courses in March!

👉https://physalia-courses.org/courses-workshops/
#Genomics #MetaAnalysis #Rstats #brms #GAMs #Proteomics #MassSpectrometry #NetworkAnalysis @Amandatron89 @nanopore @lgatto @wiernik @gavinsimpson

Very happy to announce that the third edition of the #BayesHSC school will be held from June 5th to June 9th in Verona, Italy!

And I will be teaching an introduction to Bayesian linear models.

Info here: https://sites.hss.univr.it/bayeshsc

#Bayesian #rstats #brms

Very happy to announce that the third edition of the #BayesHSC school will be held from June 5th to June 9th in Verona, Italy!

And I will be teaching an introduction to Bayesian linear models.

Info here: https://sites.hss.univr.it/bayeshsc

#Bayesian #rstats #brms

@wviechtb @mccarthymg should probably do that for switching from #BRMS to #PyMC 😬