I've thinking for some time about entering into causal inference, but I'm not sure how useful it can be for physics. I guess it's really helpful for social sciences and health sciences, but since in physics causality and causes are well established and known (at least in classical physics), I have doubts.

Are you people in #physics using #causalinference? #statistics

Hello sciences.social! I work on social policy and programme evaluations, primarily RCTs and quasi-experiments, at the National Centre for Social Research (NatCen) in London, UK.

#introduction #evaluation #ProgramEvaluation #PolicyEvaluation #SocialScience #ProgrammeEvaluation #CausalInference #RStats #DAGs #methodology #RCT #QuasiExperiment

New overview paper in Annual Reviews by Jennie Brand, Xiang Zhou, and Yu Xie on "Recent Developments in
Causal Inference and
Machine Learning" https://www.annualreviews.org/doi/pdf/10.1146/annurev-soc-030420-015345 #causalinference #causality #machinelearning #sociology

My wonderful post-doc, health #economist Julia Dennett talks on her paper on the long-term effects of #OxyContin marketing in America. #harmreduction #causalinference #epidemiology #publichealth

https://www.healthaffairs.org/do/10.1377/hp20230731.712412/full/

Our new paper on the long-term impact of opioid manufacturers targeted marketing to docs in states with high cancer burdens in the 1990s. 25 years later: more deaths from hepatitis and infective endocarditis. #opioids #Purdue #OcyContin #HCV #HBV #endocarditis #epidemiology #causalinference

https://www.healthaffairs.org/doi/10.1377/hlthaff.2023.00146

Random Tuesday evening thought... Has any good #causalinference #research been done on the #health effects of council leisure centres in the UK? Because honestly, nothing makes me more of patriotic semi-socialist Karen. Cut my local library and I'll grumble, but come after the council swimming pool where I had birthday parties as a kid, and I'll be stacking furniture into barricades.

New accepted paper with Shree Vallabha and Felicity Turner-Zwinkels!

The onset of COVID-19 made people feel stressed and threatened, but *did not cause much attitude change*

The change that was found, wasn't predicted by dominant theories.

Lots of robustness checks & the conclusions hold

Full text: https://osf.io/ahtk6

#COVID19 #polisci #socpsych #research #causalinference @socialpsych @politicalscience

6/

In a more recent work [5], the authors tested LLMs on *pure* #causalInference tasks, where all variables are now symbolic (screenshot 1). They constructed systematically a dataset starting by picking variables, to generating all possible #causalGraphs, to finally mapping all possible statistical #correlations. They then “verbalize” these graphs into problems for LLMs to solve for a given causation hypothesis (screenshot 2).

#Paper #NLP #NLProc #Causation #CausalReasoning #research

I don't think I wrote a new #introduction when I migrated from .social.
I'm a #computational #InfectiousDisease #epidemiologist and #biostatistician. I work on #CausalInference and #NetworkScience methods in ID Epi and #DiseaseEcology #DisEcol

Our first workshop on Econometric Theory and Applications is approaching. We are looking forward to hosting an amazing group of researchers in Siena. Program at http://bit.ly/3jb5dYb #SWETA #UofSiena #Econometrics #CausalInference

🚨🚨🚨New Blog Post🚨🚨🚨

How I use 🎲 and gganimate to help students "feel" the riskiness of under-powered tests.

#poliscitwitter, #econtwitter, #academictwitter, #rstats, #causalinference

http://www.carlislerainey.com/blog/2023-06-01-teaching-confidence-intervals-with-gganimate/

very interesting paper on causal inference methods https://muse.jhu.edu/article/895650 #causality #causalinference #causalfriday

Central to this work is observational data analysis properly controling for confounding to get clean causal effects. Frankly, #Ecology isn't great at #CausalInference w/ observational data. We've been all about experiments since the 1960s, missing developments in other fields.

Check out my new, free, open source software for extracting causal DAGs from text. (with or without LLM)
#causalinference #LLM
https://qbnets.wordpress.com/2023/04/24/welcome-to-the-first-public-version-of-mappa-mundi/

What’s your go-to reference for best practices for power analysis for a survey experiment? Maybe I’m not up-to-date, but it seems like a big gap between the well-developed theory and the (missing?) practical, concrete advice.

#poliscitwitter #econtwitter #academictwitter #rstats #causalinference

Applications are rolling in! The Yale Department of Epidemiology of Microbial Diseases is looking for two new tenure track faculty. Yale EMD is a great place to work with wonderful colleagues and students. Please spread the word.

https://apply.interfolio.com/122824

#epidemiology #epiverse #causalinference #publichealth #yale #ysph

https://apply.interfolio.com/122824

Excited that MIT Technology Review is covering our Nature Portfolio Machine Intelligence paper co-authored with @BernhardKainz and Ciarán M. Gilligan Lee

Check out how we employ counterfactual reasoning to empower decision making in a wide range of fields, 👇

https://www.technologyreview.com/2023/04/04/1070885/complex-math-counterfactuals-spotify-predict-finance-healthcare/

#machinelearning #causality #causalinference #econometrics

Does anyone have any readings about evaluating the quality of matches when using a matching approach to causal inference (ideally k-nearest neighbor w/ Mahalonobis distance, not PSM)? How to double-check that your matching is at least somewhat adequately accounting for confounders that go along with self-selection into the treatment? #causalinference #impactevaluation #statistics

BONUS for some kind of placebo test - match a subset of the control pop. (a "pseudo-treatment") to other controls.

Extraction of causal DAGs from text using DALL-E
#causalinference #causality #causalai
https://qbnets.wordpress.com/2023/03/28/how-to-use-dall-e-to-create-a-comic-book-from-a-short-story-or-novel/

📢On May 26th, we (the ODISSEI Social Data Science team) will do a workshop on methods for causal impact assessment for social scientists!

📈We will build #causalinference intuition with potential outcomes and immediately put this in practice using #rstats :rstats:

🔗More info + registration here:
https://odissei-data.nl/en/2023/03/soda-team-workshop-causal-impact-assessment/