Christian Bartels

Single-world intervention graphs (SWIGs) as distributions. Conceptually, simultaneously related to and distinct from Rubin's framework and Pearl's calculus. doi.org/10.48550/arX... #SWIG #DAG #causal #identification #frontdoor #docalculus

Excited to share that my paper with Zhongren Chen on generalizing causal effects in the presence of non-compliance is now out in Political Analysis! doi.org/10.1017/pan....

The data generating process: something that does not exist. Invented by scientists to earn their living.

+ + BREAKING ++ Switzerland votes in one year. Latest polls suggest the radical right SVP could surpass its 2015 record, approaching 30%. But rather than framing this as a surge, the story of the SVP: a relatively stable electoral base, alongside an increasing number of defeats at the ballot box

Why did 2 large, pivotal clinical trials [vs interleukin-6 and Lp(a)] fail to lower heart attacks, strokes and cardiovascular deaths? erictopol.substack.com/p/the-corona...

Uruguay, with 3.5 million people, generates 98% of its electricity from renewable sources. The architect of that transformation, Ramón Méndez Galain, put it simply: "The key is not technology; it is institutions. Once the rules are fair and predictable, the system builds itself."

🚨 Big news 🚨 After 10 years at Harvard, my lab is moving to Barcelona in Spring 2027 to join @cbc-upf.bsky.social and the Department of Engineering at @upf.edu. 1/4

Causal inference without do-calculus nor potential outcomes. An important simplification? #causal #causal_inference #swig #docalculus #identification

Coding agents like Claude Code feel like magic but they’re really a bunch of tools in a trenchcoat. This week I build a working coding agent in R, starting from just three tools. I’ll add some extra tools to make it safer and more efficient. tidydesign.substack.com/p/a-coding-a...

Many of us can save a child's life, if we rely on the best data. Giving money to a charity is one of the best things you can do for others. Your donation can make a huge difference. But whether or not your donation makes a difference greatly depends on *where* you donate.

A population cap would disrupt Switzerland’s relationship with Europe. Register for free to read how the country should instead regulate the inflow of migrants without breaking EU agreements

Early us of Bayesian approach in bioinformatics. Use available knowledge on chemical shifts and distances in homologous proteins as prior to assign NMR spectra for protein structure determination. en.wikipedia.org/wiki/Structu... doi.org/10.1007%2FBF... #nmr #protein #structure #bayesian

Early use of directed acyclic graphs to speed up analysis of mass spectrometry of peptides. Low polynomial time complexity, down from exponential complexity without DAGs. en.wikipedia.org/wiki/De_novo... doi.org/10.1002/bms.... #DAG #Peptide #sequencing

It's alive! 🎉 𝗧𝗵𝗲 𝗗𝗮𝘁𝗮 𝗔𝗻𝗮𝗹𝘆𝘀𝘁'𝘀 𝗚𝘂𝗶𝗱𝗲 𝘁𝗼 𝗖𝗮𝘂𝘀𝗲 𝗮𝗻𝗱 𝗘𝗳𝗳𝗲𝗰𝘁 is out -- an introduction to causal inference in practice. The first two chapters are available for free here: theissbendixen.com/dag-book/ More below 👇

In unsicheren Zeiten brauchen wir stabile Beziehungen zu Europa. Aber die Chaos-Initiative reisst genau diese Brücken ein – sie ist politisch und wirtschaftlich ein Eigentor, das unsere Sicherheit und unseren Wohlstand gefährdet.

Single-world intervention graphs (SWIGs) as distributions. Systematic way to derive identifying expressions for estimands. New front-door derivations extending more readily to complex settings. doi.org/10.48550/arX... #SWIG #DAG #causal #identification #frontdoor

Everything is a modeling problem.

Finally uploaded a revised version of this preprint on parameter identifiability arxiv.org/abs/2502.04867. Approach called 'invariant image reparameterisation' though 'image' used in mathematical sense not picture sense. Took me a while to revise for various reasons, but fairly happy with it now

Finally switched to DuckDuckGo and it’s nice to not have LLM summaries anymore

How will populations across the world change in the 21st century? 🔧 Explore for yourself with our new interactive tool!

One week away! Don't miss the next Methods Series presentation with Alex Ocampo.

The median age in China has rapidly caught up with the United Kingdom— In 1965, the median age in the United Kingdom was almost twice that of China. Half of the people in the UK were younger than 34 years, and half were older. In China, this midpoint was just 18 years.

least sexist industriebranche

Meanwhile, Spain's massive investment in renewables is paying dividends now: with prices for Spanish industry and consumers low and stable compared with other European economies. www.ft.com/content/ac77...

Post a pic you took, no context, to bring some zen to the feed.

spending my sunday evening once again attempting to draw a DAG for diff-in-diff

📣The 2026 Symposium of #CausalInference in the #HealthSciences takes place on March 18, 2026 in Fribourg. Theme: AI & machine learning in causal inference for health sciences 🎤Keynotes: Elsa Gautrain, Aurélien Sallin, Jonas Peters, Jana Mareckova 🔗https://projects.unifr.ch/pophealthlab/?page_id=1561

We apply the estimand framework to dose–exposure–response analyses. ... strategy to improve exposure–response analyses for dose selection, particularly when the relevant evidence includes data from multiple studies. #estimand #exposure-response #dose-response #causal doi.org/10.1002/psp4...

For years I had trouble following some of the discussion about confidence bands, but at ACIC this year @noahgreifer.bsky.social pointed me to a helpful paper So you don't have to be as perplexed as I once was, we have a new pre-print introducing the key ideas arxiv.org/abs/2510.07076

I'm so excited to announce the first release of my newest #Rstats package, {adrftools}! This package facilitates estimation, visualization, and testing for the causal effect of a continuous (i.e., non-discrete) treatment. 🧵 1/10 #statssky #episky #causalinference

As I said, this is the important work that needs to be done. Mark's paper is overly simplistic, arguing we can't judge deviations at all. Deviations are in practice not remotely as bad as he wants. If he had collected actual data,che would have falsified his own claims.

This is a fairly technical but highly relevant paper on how we can model complex systems at various levels of detail without losing causal content. Think gas: instead of tracking every molecule, we can focus on big-picture properties like temperature and pressure. www.auai.org/uai2017/proc...

What are some of the best DAGs you seen that depict time-varying confounding? Could be from a 'real' worked example, or generic. #EpiSky

arXiv📈🤖 Long-Term Causal Inference with Many Noisy Proxies By Lal, Imbens, Hull

“Uruguay did what most nations still call impossible: it built a power grid that runs almost entirely on renewables—at half the cost of fossil fuels. The physicist who led that transformation says the same playbook could work anywhere—if governments have the courage to change the rules.”

X/Twitter's rough full volume is around 500 total million posts every day, or 182 (and a half) billion posts per year. By contracts, we found 11.2 million research posts in all of 2025 on there. In other words, 0.000006% of Twitter appears to be sharing research. Basically zero.

This gives financial institutions leeway to acquire hard-to-sell, higher-yielding long-term assets and finance them with cheaper short-term liabilities, thus increasing profits through a risky “mismatched book”.

link 📈🤖 Demystifying Proximal Causal Inference (Ringlein, Nguyen, Zandi et al) Proximal causal inference (PCI) has emerged as a promising framework for identifying and estimating causal effects in the presence of unobserved confounders. While many traditional causal inference methods rely on the

Happy new year! As an idea for improved privacy pr.tn/ref/X11M0Y8Q

Lumo now understands LaTeX. Type any formula (fractions, matrices, integrals, etc.) straight into chat and see it rendered instantly. No more screenshots or ASCII tricks. Perfect for precise technical work and academic tasks. Try it now: https://lumo.proton.me

Students learn early that correlation does not imply causation. Correct, but incomplete... Certain patterns of independence and conditional dependence constrain #causal structure very strongly. A simple example:

IL-17-mediated antifungal immunity restricts Candida albicans pathogenicity in the oral cavity www.nature.com/articles/s41... The microbiome not only consists of bacteria, but also of fungi. Most of them support human and animal health medicalxpress.com/news/2025-12... However, some fungi ...

Despite claiming to have seized full control of Pokrovsk, it is not clear that Russia has done so—or that it can necessarily press on for longer than Ukraine can hold on. Our charts break down the data