Martin Modrák
LightNews LightNews
Martin Modrák
@modrakm.bsky.social
1.3K followers 36 following 170 posts
Biostatistics/bioinformatics at Charles University, 2nd faculty of Medicine. Bayesian in practice, but not a fan of Bayesian epistemology. Main on fedi: https://bayes.club/@modrak_m
Blog: https://martinmodrak.cz
Posts Replies Media Videos
Reposted by Martin Modrák
konsta.happonen.eu
Lighting the Bayesian bat signal: #stan #rstats #statssky

A latent construct C is informed by multiple variables X like so:

X₁, X₂ ~ C

I want to use C to also predict Y,

Y ~ C

but without C being informed by Y. What's the cleanest way to implement this in Stan?
February 7, 2026 at 9:49 AM
Reposted by Martin Modrák
emilhvitfeldt.bsky.social
I'm happy to announce a new and improved version of my quarto revealjs extension that interfaces with roughnotation

More annotation types, fragment support, works with flourish extension to allow for annotations of code chunks, and reverse animations are all here

github.com/EmilHvitfeld...
#quarto
February 5, 2026 at 8:14 PM
Reposted by Martin Modrák
rmcelreath.bsky.social
I am doing a short workshop in this symposium (quoted below) for folks in and around Leipzig. Here's the abstract. I really indulged myself with this one.

A Guerilla Approach to Scientific Workflow:
Title: A Guerilla Approach to Scientific Workflow Abstract: Taking inspiration from guerilla warfare, I will introduce a framework for managing scientific research that addresses the complexity of research while advancing goals of transparency and reliability. Real science is complex, and there is no single scientific method. It is a chaotic network of theories, data, models, graphs, comparisons, summaries, and narratives. A flaw in any part can ruin everything. How can small teams of researchers challenge this complexity and emerge victorious? More statistical training is not enough. More resources is not enough. Lawrence of Arabia once said, "Guerrilla warfare is more intellectual than a bayonet charge." And this is how we too will proceed. Instead of confronting the complexity directly, with more money and more people and more data and more meta-analyses, we must develop a SCIENTIFIC WORKFLOW that subdivides, analyzes, and transparently justifies our individual research projects. The tools to do this already exist, and individual researchers can become scientific guerillas today to improve their chances of victory.
February 4, 2026 at 12:05 PM
Reposted by Martin Modrák
richarddmorey.bsky.social
As far as I'm aware, this is the first time anyone has shown that innocuous data-dependent reporting practices can bias meta-analyses (please let know if I'm wrong!), so we should be very careful interpreting big metascientific meta-analyses going forward. 9/x
February 2, 2026 at 2:56 PM
Reposted by Martin Modrák
statsepi.bsky.social
Hello friends. I'm looking for an existing SOP describing how CTUs deal with database lock when a blinded RCT is followed by an open label extension study where all patients can be offered the active tx once their planned follow-up in the RCT completes. Thanks! #clinicaltrials
February 2, 2026 at 8:34 PM
Reposted by Martin Modrák
richarddmorey.bsky.social
New draft: "Decline effects, statistical artifacts, and a meta-analytic paradox". In this manuscript I show how a common practice in meta-analysis (eg the 2015 Open Science Collaboration) creates artifactual signatures of poor scientific behavior. PDF: raw.githubusercontent.com/richarddmore... 1/x
The decline effect (Protzko & Schooler, 2017) is an observed phenomenon where effect sizes in experiments apparently diminish in size from the first paper demonstrating the effect to later replications. This has been taken as a symptom of an unhealthy scientific ecosystem, possibly caused by the "winner's curse" (selection on significance and regression to the mean), publication bias or opportunistic analyses. I show that decline effects can arise as an artifact from a much simpler source: the original article determining the sign of the effect in a meta-analysis. Moreover, such artifactual decline effects will show correlations with some of the same experimental properties that one would expect from biases from poor behavior, such as the sample size of the original study.
February 2, 2026 at 2:56 PM
Reposted by Martin Modrák
aurelienallard.bsky.social
It's a beautiful plot, but it's terribly misleading about the impact of pre-registration. More recent studies (with higher sample sizes) find very little impact of pre-registration on the publication of null results.

Here's a thread with some references (1/N)
lakens.bsky.social Daniel Lakens @lakens.bsky.social · 9d
@scientificdiscovery.dev Hi, I am creating a new version of my free online MOOC, and would like to use this picture, of which you have the copyright. Is it ok if I put it in a slide?
February 2, 2026 at 1:47 PM
Reposted by Martin Modrák
solomonkurz.bsky.social
#rstats bat signal:

If any of y'all have experience with multilevel multinomial logistic regression, with {brms}, Kruschke and I could use your help. This is your chance to influence the upcomming textbook. discourse.mc-stan.org/t/problem-in...
Problem in multilevel (hierarchical) multinomial logistic regression (with brms)
I have posted a teaser of the issue at my blog, copied below. The data The predicted variable is a categorical response, named resp with levels ‘1’, ‘2’, ‘3’, and ‘4’ (nominal labels, not numerical...
discourse.mc-stan.org
January 30, 2026 at 3:02 PM
Reposted by Martin Modrák
avehtari.bsky.social
Bayesian Workflow by
Andrew Gelman, Aki Vehtari, @rmcelreath.bsky.social with @danpsimpson.bsky.social, @charlesm993.bsky.social, @yulingy.bsky.social, Lauren Kennedy, Jonah Gabry, @paulbuerkner.com, @modrakm.bsky.social, @vianeylb.bsky.social

(in production, estimated copy-editing time 6 weeks)
**Part 1: From Bayesian inference to Bayesian workflow** 1. Bayesian theory and Bayesian practice 2. Statistical modeling and workflow 3. Computational tools 4. Introduction to workflow: Modeling performance on a multiple choice exam **Part 2: Statistical workflow** 5. Building statistical models 6. Using simulations to capture uncertainty 7. Prediction, generalization, and causal inference 8. Visualizing and checking fitted models 9. Comparing and improving models 10. Statistical inference and scientific inference **Part 3: Computational workflow** 11. Fitting statistical models 12. Diagnosing and fixing problems with fitting 13. Approximate algorithms and approximate models 14. Simulation-based calibration checking 15. Statistical modeling as software development **4. Case studies** 16. Coding a series of models: Simulated data of movie ratings 17. Prior specification for regression models: Reanalysis of a sleep study 18. Predictive model checking and comparison: Clinical trial 19. Building up to a hierarchical model: Coronavirus testing 20. Using a fitted model for decision analysis: Mixture model for time series competition 21. Posterior predictive checking: Stochastic learning in dogs 22. Incremental development and testing: Black cat adoptions 23. Debugging a model: World Cup football 24. Leave-one-out cross validation model checking and comparison: Roaches 25. Model building and expansion: Golf putting 26. Model building with latent variables: Markov models for animal movement 27. Model building: Time-series decomposition for birthdays 28. Models for regression coefficients and variable selection: Student grades 29. Sampling problems with latent variables: No vehicles in the park 30. Challenge of multimodality: Differential equation for planetary motion 31. Simulation-based calibration checking in model development workflow **Appendices** A. Statistical and computational workflow for Bayesians and non-Bayesians B. How to get the most out of Bayesian Data Analysis
January 26, 2026 at 8:18 AM
Reposted by Martin Modrák
jessicahullman.bsky.social
Wrote a summary of a great keynote by @zey.bsky.social at NeurIPS, arguing that we’re having the wrong nightmares about AI: not AGI or superhuman benchmarks, but good-enough genAI at scale threatens "load bearing frictions" society relies on to signal effort, authenticity, sincerity, credibility.
Zeynep Tufecki on having the wrong nightmares about generative AI
I was writing a blog post where I was going to reference Zeynep Tufecki’s 2025 NeurIPS keynote, and realized there isn’t a solid synopsis online.
open.substack.com
January 9, 2026 at 4:42 PM
Reposted by Martin Modrák
cedricscherer.com
I ignored the strip.clip argument in #ggplot2 for way too long 😲

Combined with a small negative margin tweak, you can place facet labels inside each panel. A tiny trick that makes small multiples feel so much cleaner.

🔵 no manual coordinates
🔵 inherits theme styling
🔵 scales nicely when resizing
Side-by-side comparison of two multi-panel bubble charts faceted by world region. The left column shows the default facet labels placed above each panel (“Africa”, “Americas”, “Asia”, “Europe”, “Oceania”). The right column shows the same charts, but the facet labels are moved inside each panel at the top-left using a negative margin. In the center, there is a title reading “Want to place your facet labels inside each panel?” with an arrow pointing right, followed by a short ggplot2 theme code snippet demonstrating how to move strip text inside the panel.
December 12, 2025 at 12:51 PM
Reposted by Martin Modrák
bretdevereaux.bsky.social
You can see some of these stereotypes - the simple, morally pure countryside vs. the morally compromised, inauthentic city - play out in Greek and Roman literature.

So this is a very old idea that recurs regularly.
astrokatie.com Katie Mack @astrokatie.com · Jan 1
Do other countries have this weird notion that you’re not a “real” representative of the nation if you live in an urban center? Like do the French say Parisians aren’t really French? Are you considered not a real German if you live in Berlin? Or is this mainly a weird American thing?
January 1, 2026 at 7:24 PM
Reposted by Martin Modrák
statsepi.bsky.social
‪It has a name now 😜

Many thanks to Ken for agreeing to put his good name to my...artwork. The image is in the public domain (CC 0), but citations to the linked documents are warmly welcomed.

✅ zenodo.org/records/1808...

✅ pubmed.ncbi.nlm.nih.gov/24452418/
Figure 1: The Rothman-Dahly Evidence Pyramid (original version) An equilateral triangle with a small blue section labelled "Thoughtful, well-conducted studies of any design" at the top, with the remaining space colored red and labelled "The other shit"
December 29, 2025 at 11:19 AM
Reposted by Martin Modrák
I'd like to propose the following norm for peer review of papers. If a paper shows clear signs of LLM-generated errors that were not detected by the author, the paper should be immediately rejected. My reasoning: 1/ #ResearchIntegrity
December 28, 2025 at 6:23 AM
Reposted by Martin Modrák
astrezh.bsky.social
We have weirdly benefitted from the fact that all the 2010s data science boom Medium posts littering the training corpus were so consistently terrible.
December 18, 2025 at 2:10 AM
Reposted by Martin Modrák
StanCon 2026 registration and abstract submission are now open
statmodeling.stat.columbia.edu/2025/12/11/s...
StanCon 2026 registration and abstract submission are now open | Statistical Modeling, Causal Inference, and Social Science
statmodeling.stat.columbia.edu
December 11, 2025 at 1:46 PM
Reposted by Martin Modrák
mjskay.com
This could be very useful for R -> InkScape and R -> Illustrator workflows! #rstats

See the vignette: cran.r-project.org/web/packages...
December 11, 2025 at 2:50 PM
Reposted by Martin Modrák
rmcelreath.bsky.social
I'm teaching Statistical Rethinking again starting Jan 2026. This time with live lectures, divided into Beginner and Experienced sections. Will be a lot more work for me, but I hope much better for students.

I will record lectures & all will be found at this link: github.com/rmcelreath/s...
course schedule as a table. Available at the link in the post.
December 9, 2025 at 1:58 PM
You will be visited by 3 spirits
An RStudio console with output of a Stan run. It ends in 3 warnings: 1: There were 38 divergent transitions after warmup. 2: There were 836 transitions after warmup that exceeded the maximum treedepth. 3: Examine the pairs() plot to diagnose sampling problems
December 9, 2025 at 9:36 PM
Reposted by Martin Modrák
phdemetri.bsky.social
You will be visited by 3 spirits
December 9, 2025 at 4:29 PM
Reposted by Martin Modrák
devezer.bsky.social
as awful as this is, these examples kinda make me feel giddy and validated, given how i normally feel about our current peer review practices. my personal experience unfortunately has never been consistent with academics' general regard of peer review as a net benefit despite its flaws.
carlbergstrom.com Carl T. Bergstrom @carlbergstrom.com · Nov 28
Springer-Nature statement

“Whilst the details of peer review are confidential, we can confirm that the article underwent two rounds of review from two independent peer reviewers, supporting an accept decision.”

How am I now expected to believe that two people looked at the paper twice and DGAF?
Riding the Autism Bicycle to Retraction Town
Does anyone *really* know their Factor Fexcectorn?
nobreakthroughs.substack.com
November 28, 2025 at 5:14 PM
Reposted by Martin Modrák
arrianna-planey.bsky.social
This touches on a few of the biggest sources of low morale for me:

- flood of AI slop lands in my inbox as peer review requests
- my critical reviews of said AI slop are then ignored by editors at Journals that won't give my work consideration
- all while I'm expected to ⬆️ research output
jessicacalarco.com Jess Calarco @jessicacalarco.com · Nov 28
It's easy to see shoddy research as a bad actor problem. But if AI slop like this can make it through editors and peer reviewers, it means there are systemic problems at work. And I'd argue that at least part of the problem is the overwork culture in academia-- pressure to do more while caring less.
erikangner.com Erik Angner @erikangner.com · Nov 27
"Runctitiononal features"? "Medical fymblal"? "1 Tol Line storee"? This gets worse the longer you look at it. But it's got to be good, because it was published in Nature Scientific Reports last week: www.nature.com/articles/s41... h/t @asa.tsbalans.se
November 28, 2025 at 5:43 PM
Reposted by Martin Modrák
pengzell.bsky.social
A thread of directed acyclic graphs (DAGs) that look like record covers... because that's EXACTLY what the world needs

1. Huey Lewis and the News: link.springer.com/chapter/10.1...
November 28, 2025 at 11:36 AM
Reposted by Martin Modrák
bapalmer.bsky.social
Whenever I see campaigns such as this, I'm reminded of the 2021 tweet from Surrey Poilce
November 27, 2025 at 9:37 AM
Reposted by Martin Modrák
richarddmorey.bsky.social
Anyone have experience analysing periodicity (eg with Fourier) in data with gaps/missings (eg weekends, holidays) - this is in a health context if it helps? Looking for guides to pitfalls, etc
November 26, 2025 at 12:00 PM