jepusto · @jepusto
326 followers · 162 posts · Server fediscience.org

Okay Stats-o-don, show me that you're more alive than the (formerly)🦜site. Please hit me with favorite refs/guidance on analysis of randomized trials where Tx is assigned within sites for part of the sample but between sites for the other part (i.e., hybrid of individual- and cluster-randomization).

#CausalInference #rct #stats

Last updated 2 years ago

Andrew Heiss :rstats: · @andrew
3855 followers · 1531 posts · Server fediscience.org

Added a new laptop sticker: @epiellie's "Eat Ice Cream; Do Crime" DAG showing the sun as confounder for ice cream and crime (get it and other causal inferencey things from redbubble.com/people/epiellie/)

#CausalInference

Last updated 2 years ago

Andrew Heiss :rstats: · @andrew
3762 followers · 1396 posts · Server fediscience.org

throwing some do() operators into this conjoint blog post so you know it's serious causal inference

#statsodon #CausalInference

Last updated 2 years ago

Miguel Hernan · @MiguelHernan
1150 followers · 42 posts · Server fediscience.org

>66,000 individuals provided info at their doctor’s office or in their homes

4400 health professionals in >1400 primary care centers were mobilized

An information system for 2000 concurrent users was created

Agencies of the Spanish government and 17 regions were coordinated

No fancy methods were used.

ajph.aphapublications.org/doi/

Respect to colleagues who launched this study in less than 4 weeks in a health system under pandemic pressure.

Respect to descriptive studies.

#CausalInference

Last updated 2 years ago

Miguel Hernan · @MiguelHernan
1150 followers · 40 posts · Server fediscience.org

We talk much about , but is the foundation of everything.

Descriptive methods aren't as glamorous as causal inference methods but, without good description, causal inference is toast.

We can only make sound causal inferences if we can adequately *describe* the treatments, outcomes, confounders, eligibility criteria...

Good description requires inspecting and validating info in existing databases or, when not available, generating the data from scratch.

#description #CausalInference

Last updated 2 years ago

ai-jobs.net · @aijobs
37 followers · 163 posts · Server mstdn.social
ai-jobs.net · @aijobs
37 followers · 154 posts · Server mstdn.social
ai-jobs.net · @aijobs
37 followers · 154 posts · Server mstdn.social
ai-jobs.net · @aijobs
37 followers · 154 posts · Server mstdn.social
ai-jobs.net · @aijobs
33 followers · 141 posts · Server mstdn.social
Miguel Hernan · @MiguelHernan
1138 followers · 37 posts · Server fediscience.org

On April 5, I'll be talking about for in health research.

What can we learn from data and how much learning can we automate?

Join us in person at Harvard T.H. Chan School of Public Health (Kresge 502) or online at

eventbrite.com/e/ai-for-causal

#CausalInference #ai

Last updated 2 years ago

ai-jobs.net · @aijobs
23 followers · 97 posts · Server mstdn.social

Talking about DAG colliders in my program evaluation class tonight, so back to my good ol' standby examples of niceness → appearance + race → police use of force

#CausalInference #statsodon

Last updated 3 years ago

Dr Jen Dykxhoorn · @jendykxhoorn
144 followers · 36 posts · Server fediscience.org

Fully-funded PhD post in
UCL Division of Psychiatry with
Prof. James Kirkbride and me

✨ Examining the causes of health and social inequalities in psychiatric disorders ✨

Apply now! findaphd.com/phds/projectdetai

#datascience #CausalInference #epiverse #epidemiology #phdchat

Last updated 3 years ago

Peter Tennant, PhD · @pwgtennant
1738 followers · 911 posts · Server fediscience.org

A handful of places are still available on our Introduction to Course for Health & Social Scientists (15-19 May 2023, Leeds, UK). Book *NOW* to avoid disappointment!

For more information, & to register, please visit: causal.training/

#epiverse #causalintrocourse #CausalInference

Last updated 3 years ago

Santam Chakraborty · @santam
294 followers · 1276 posts · Server fediscience.org

Seeking help from folks for appropriate analytical strategy for a study investigating the impact of diabetes on toxicities experienced by patients with cervical cancer receiving radiotherapy. discourse.datamethods.org/t/ad

#analysisplan #CausalInference #dag #analysis #statistics #radonc #cancer

Last updated 3 years ago

CAUSALab · @causalab
45 followers · 13 posts · Server mstdn.social

Exciting news!

CAUSALab researcher and Assistant Professor Andrew Beam is Co-Host of new podcast NEJM AI Grand Rounds that will be showcasing hot topics on , and .

Check out the podcast and hit follow👉 ai-podcast.nejm.org

#CausalInference #MachineLearning #artificialintelligence

Last updated 3 years ago

CAUSALab · @causalab
45 followers · 13 posts · Server mstdn.social

Time's running out! Early-bird registration for American PsychoPathological Association's 2023 Conference ends next week.

CAUSALab Director @MiguelHernan will be giving a talk on in research.

This year's conference organizer is Dost Ongur, Co-Director of LEAPCenter with @MiguelHernan.

Visit appassn.org/home/2023-annual-m to register.

#appa23 #mentalhealth #CausalInference

Last updated 3 years ago

John Kubale · @jkubale
56 followers · 22 posts · Server fediscience.org

I've tinkered around with sections of Statistical Rethinking by @rlmcelreath in the past, but this year I'm aiming to follow along through the entire course. Holy wow. It's so good.

#CausalInference #bayesian #rstats

Last updated 3 years ago

The 8th iteration of my and course is up and live at evalsp23.classes.andrewheiss.c !

It covers basic econometrics and DAGs, all with , and it's mostly asynchronous with dozens of hours of videos, and the whole thing is Creative Commons-licensed, so do whatever you want with it!

#statsodon #EconTwitter #epitwitter #rstats #CausalInference #ProgramEvaluation

Last updated 3 years ago