Want to master probabilistic modeling for real-world problems?

@allendowney.bsky.social is co-teaching Applied Bayesian Modeling with #PyMC (Nov 2 to Dec 2). 4 weeks of live, hands-on instruction directly from the creators and maintainers.

Register here: dub.sh/mhCpq6U
#Bayesian

Thermodynamic Cyclic Processes with Markov Samplers in Bayesian Inference

Heinrich von Campe, Bjoern Malte Schaefer

Action editor: Gilles Louppe

https://openreview.net/forum?id=88KWqihymD

#markov #bayesian #mcmc

Another thing I really liked from @rmcelreath.bsky.social’s talk yesterday: he introduced #BayesianWorkflow, with Andrew Gelman, Aki Vehtari et al. It’s much more than just #Bayesian inference: …

1/2

#statstab #631 Prior Information in Frequentist Research Designs: The Case of Neyman’s Sampling Theory

Thoughts: Model-based v design-based problem. This should be discussed more.

#modelbased #designbased #sampling #priors #neyman #frequentist #bayesian

link.springer.com/article/10.1...

In chats with collaborators over the past few days, I've seen how poorly generative AI proposes and evaluates Bayesian models. It really struggles with hierarchical modeling and spotting nonidentifiability. Was its training supervised mostly by frequentists? 😉
#Bayesian #AI

#statstab #626 Goal-Driven Flexible Bayesian Design

Thoughts: Bayes allows you to design informative studies and anticipate issues.

#bayesian #bayes #design #simulation #rct #tutorial #inference #mcid #equivalence #efficacy #estimation #noninferiority

hbiostat.org/bayes/design

How can we make Bayesian ANOVA models truly reusable? 🤔Instead of rewriting JAGS code for every experimental design, we can leverage matrix notation (Y = X β) alongside R’s model.matrix() and emmeans().
Read the full post on www.statforbiology.com/posts/bayesi...
#Rstats #Bayesian #JAGS

#BJA editorial by Sidebotham explaining why #anaesthesia research should rely on #Bayesian probabilities, rather than #frequentist probabilities, to achieve clinically meaningful #inference statistics.

https://www.bjanaesthesia.org/article/S0007-0912(26)00345-4/fulltext

Bayesian A/B testing gains traction in commercial experimentation platforms like GrowthBook, LaunchDarkly, and more. But are the benefits oversold? A deep dive reveals that Bayesian and frequentist approaches are more similar than widely believed. #A/Btesting #Bayesian #Frequentist

#statstab #621 A General Method For Estimating Reliability Using Bayesian Measurement Uncertainty

Thoughts: "estimating reliability [...] for complex cognitive and behavioural assessments without test-retest data" wow!

#reliability #bayesian #testretest

doi.org/10.31234/osf...

What if we learned the likelihood instead of the posterior?

In Episode 165, Alex Fengler joins @alex-andorra.bsky.social to discuss HSSM, likelihood approximation networks, amortized inference and simulation-based inference and more ...

🔗 learnbayesstats.com/episode/hssm...

#Bayesian #PyMC

New paper by @masahironakano.bsky.social , @casewell.bsky.social & @clopathlab.bsky.social who show that planning-like L/R #ThetaSequences in #hippocampus can arise from #MEC dynamics and #Bayesian decoding. Take home: Not every future-like sweep is necessarily planning.

#Neuroscience #CompNeuro

#statstab #616 The brms Book: Applied Bayesian Regression
Modelling Using R and Stan (Early Draft)

Thoughts: {brms} is a main package fir bayesian stats. A must to know.

#brms #bayes #bayesian #modelling #book #tutorial #regression #stan #rstats #r

paulbuerkner.com/software/brm...

'Phylochronology' is my word of the day 🏺 This paper ticks two research interests for me:
Ancient DNA #aDNA, and #Chronology construction using #Bayesian techniques to combine multiple dating sources.
doi.org/10.1073/pnas...

New study out in JSP!

Dating the origin age of earwigs: impact of the earliest record of stem-group Dermaptera (Insecta: Protelytroptera) based on Bayesian inferences from the fossil record by Peng et al. Read here: buff.ly/SnX92HQ

#Bayesian #Insecta #Polyneoptera #Protelytroptera #PaleoSky

#statstab #611 How to practise Bayesian statistics outside the
Bayesian church: What philosophy for Bayesian statistical modelling?

Thoughts: A comment on the last statstab on the philosophy of Bayesian.

#philosophy #bayes #inference #bayesian

bpspsychub.onlinelibrary.wiley.com/doi/epdf/10....

Is your MMM giving paid search credit for demand that TV actually created?

We show how modeling search as a mediator, not just another channel, can change attribution and lead to very different budget decisions.

🔗 Read the full article: dub.sh/j4NHQoW

#Bayesian #PyMC #MarketingAnalytics #PyMCLabs

Final call! Our #Bayesian Marketing Analytics course starts Monday, Sept 7.

Learn live with the team behind #PyMC, PyMC-Marketing, & CausalPy:

- Interactive coding (MMM & CLV)
- Bring your own data
- Private Discord Q&A

Register: dub.sh/CAtxlsn

#MarketingAnalytics