(1/22) Over the last few days I have been discussing the role of formal causal inference in RCTs, and in general the role of formalism and rigor in statistics in twitter with @f2harrell.bsky.social. Thread with an example for why both are fundamental for biomedical research:

The critique of unmeasured confounding is often levied in a lazy/broad way. It is trivially true in any observational study. But if the critic can't think of a plausible such confounder and posit a reasonable direction/magnitude of its bias then they're not doing productive science.

Arman Oganisian@stablemarkets.bsky.social · 8mo ago

We really do need to get away from the binary. It should be "is there unmeasured confounding or not." Almost certainty there is - it's a matter of how much and what direction.

New preprint out on a way to handle structural and practical violations of the overlap (also known as positivity) assumption in causal inference -- as long as the outcome is bounded, we derive simple partial identification bounds on the ATE. With @alecmcclean.bsky.social and @idiaz.bsky.social

He did it before Double Machine Learning I met with professor Mark van der Laan because I think his work is pretty incredible and it sometimes feels like a secret that only a few people know about, especially in industry. 1/ #CausalSky #StatSky #CausalInference

I have a new paper out on a simple way to do causal inference with left-censored outcomes. This comes up with environmental data because measurements often have a lower limit of detection -- e.g. a chemical is undetectable below a certain level www.tandfonline.com/doi/full/10....

Non-parametric treatment effect bounds for left-censored outcomes: estimating the effect of herbicide use on 2,4-D exposure

Causal inference is concerned with defining and estimating the effect of a exposure on an outcome. For example, the Average Treatment Effect (ATE), a causal inference concept, is defined as the pop...

tandfonline.com

I wrote something about statistics under authoritarianism

The Guardian@theguardian.com · last yr.

Why Trump’s undermining of US statistics is so dangerous | Daniel Malinsky

Underlying this there is a valid and worrisome criticism of causal inference in practice, but most comments criticizing CI as a field miss the fact that “x methodology is being abused in practice” can be correctly said about almost anything.

Darren Dahly@statsepi.bsky.social · last yr.

We didn't randomize, and there was no allocation concealment or blinding, and we can't really be sure what intervention they got or how the outcomes were measured, but we emulated a trial by drawing a DAG.

This is why I prefer causal assumptions in terms of exogenous vars in structural causal models rather potential outcomes. Sure, the former is often mathematically stronger, but the latter is inscrutable by subject matter experts.

Paul Hünermund@p-hunermund.com · last yr.

💡A new paper by Elias Bareinboim and Drago Plecko underscores the intractability of ignorability assumptions commonly invoked in the potential outcomes framework, explains why structural causal models—explicitly grounded in well-defined causal mechanisms—are far easier to interpret. 1/2

Excited to present on Thursday @eurocim.bsky.social on new work with @idiaz.bsky.social on (smooth) trimming with longitudinal data! "Longitudinal trimming and smooth trimming with flip and S-flip interventions" Prelim draft: alecmcclean.github.io/files/LSTTEs...

1/ If you were taught to test for proportional hazards, talk to your teacher. The proportional hazards assumption is implausible in most #randomized and #observational studies because the hazard ratios aren't expected to be constant during the follow-up. So "testing" is futile. But there is more 👇

Colombia. With two o’s.

From twitter: A short thread: It amazes me how many crucial ideas underlying now-popular semiparametrics (aka doubly robust parameter/functional estimation / TMLE / double/debiased/orthogonal ML etc etc) were first proposed many decades ago. I think this is widely under-appreciated!

📢📢The 4th Lifetime Data Science Conference will take place May 28–30, 2025, at New York Marriott at the Brooklyn Bridge in Brooklyn, NY, USA. This event will feature keynotes by Drs. Nicholas Jewell and Mei-Ling Lee, short courses, 60+ invited sessions, and a banquet on May 29. Register and join us!

ASA Community

The ASA Community is an online gateway for member collaboration and connection.

community.amstat.org

Happy to announce some new work with my student Kaitlyn Lee! arxiv.org/abs/2501.04871 If you're not in the know, Riesz regression is a general tool to estimate things like propensity weights without actually having to know that they are propensity weights in the first place.

RieszBoost: Gradient Boosting for Riesz Regression

Answering causal questions often involves estimating linear functionals of conditional expectations, such as the average treatment effect or the effect of a longitudinal modified treatment policy. By ...

arxiv.org

Totally agree with this, and would double down: description of causal mechanisms is the foundation of science. If we can’t describe causal mechanisms, no interventions can follow.

Miguel Hernan@miguelhernan.org · 3y ago

Don't let the causal inference buzz fool you: Description is the foundation of science. We've described the 3-year health impact of #COVID19 in Madrid, the EU region with the highest life expectancy. If we can't describe, no causal inference can follow. academic.oup.com/ofid/article...

I think it is a mistake to call one-step type estimators “debiased”. They are generally biased in a traditional sense. The problem that one-step type estimators address isn’t just about bias but more importantly about controlling the statistical behavior of the error defined as estimate minus truth.

alex hayes@alexpghayes.com · 2y ago

Isn't the whole point of the causal machine learning literature that these plug-in estimates are biased? For instance Section 4.1 of arxiv.org/abs/2203.06469

OK I’ll bite, what’s stata? The plural of statum?

Thank you Alec for leading this project, I learned a lot! This paper has a very useful study of what contrasts are feasible in situations with many treatments and positivity violations, including necessary assumptions and efficient one-step estimators. Check it out!

Alec McClean@alecmcclean.bsky.social · 2y ago

New-ish paper alert! arxiv.org/abs/2410.13522   We tackle the challenge of comparing multiple treatments when some subjects have zero prob. of receiving certain treatments. Eg, provider profiling: comparing hospitals (the “treatments”) for patient outcomes. Positivity violations are everywhere.

@wenbowu.bsky.social and I are looking for a postdoc! please reach out if you are interested.

I see renewed discussion on #statsky about the interpretation of confidence intervals. I will leave here this quote from Larry Wasserman's All of Statistics, which I love. Controlling one's lifetime proportion of studies with an interval that does not contain the parameter is surely desirable!

New paper! arxiv.org/pdf/2411.14285 Led by amazing postdoc Alex Levis: www.awlevis.com/about/ We show causal effects of new "soft" interventions are less sensitive to unmeasured confounding & study which effects are *least* sensitive to confounding -> makes new connections to optimal transport

The European Causal Inference Meeting 2025 is coming to Ghent! ✨ Share your work with experts across the globe – abstract submission for oral & poster presentations is now open! eurocim.org/abstracts.html

CMU Stats & Data Science is hiring! www.cmu.edu/dietrich/sta... forms.stat.ufl.edu/statistics-j...

Our Division is hosting its inaugural yearly Biostatistics Symposium, and this year the topic is Causal Inference! We have an exciting lineup of speakers listed below. If you are in the NYC area, please join us! Link to register in the QR below.