Social epidemiologist, wannabe statistician, Assistant Professor at NYU pop health #epitwitter
Joined April 2018
- Tweets249
- Following209
- Followers179
- Likes887
Audrey Renson retweeted
📢Join us next week, Nov. 20 at 7PM for our next teach-in, “Palestine & the Global University: Mapping Academic Repression, Scholasticide & Resistance” w/ @AKayumAhmed1, @MeiraGold, @ellisemmagarey, @thaddad
Register now: facultyforpalestine.educatio…
Have you every wondered how generalizability or transportability work with difference-in-differences? Check it out in a new preprint with @EMatthay and @kara_rudolph. arxiv.org/abs/2310.17806
Spoiler: it's wierd! You did a diff-in-diff because you had unmeasured confounding right? Well, those unmeasured confounders are probably effect modifiers too. Most of the time, you'll need to assume they're not (on relevant scale) to transport an effect from diff-in-diff.
Audrey Renson retweeted
🌟 We have the tools better understand how repeated events create health inequities, read more in our new article in @AmJEpi with @RensonAudrey @edwardsjk @iamchantelphd @LaurenZalla! Here: academic.oup.com/aje/article…
I ❤️ rigorous descriptive epi and this is a shining example. What a joy to get to work on this!
🌟New paper in @AnnalsOfEpi! We must measure inequities in life course criminal legal involvement (beyond incarceration) to understand this structural determinant of health and the harms it creates doi.org/10.1016/j.annepidem.…
🚨 Announcing my first preprint in the diff-in-diff space! A brief 🧵below. Looking for feedback/critiques!
arxiv.org/abs/2206.05788
#epitwitter #EconTwitter #causalinference
Basically, it turns out the usual diff-in-diff style assumptions can be used to learn about effects of a general time-varying treatments regime, for the whole population. Who knew???*
*Actually, @zshahn et al knew. Check out their sweet preprint about g-estimation under parallel trends! (I breathed a sigh of relief when I saw someone else independently showed identification for the same quantities) arxiv.org/abs/2204.10291
But, it's not a free lunch. Everyone has to have started out consistent with the treatment plan in question and then they can deviate over time. Also, parallel trends is kind of a weird assumption when you really look at it (scale specific, hard to use substantive knowledge).
A huge thanks to my co-authors and mentors @PausalZ, @PronouncedKeil, Michael Hudgens and Allison Aiello who are way smarter and more put-together than me!!
#craiyon a fashion show w/ dresses made out meat and hats made out of fish. I will be in attendance.
Audrey Renson retweeted
Our line up is filled with causal inference enthusiasts from around the globe:
✨Dana Goin @GoinDana
✨Audrey Renson @RensonAudrey
✨Pricila Mullachery @PricilaPhd
✨Aaron Shev
✨Michael Garber @michaeldgarber
✨Julia Lemp @jmlemp
Stay tuned for session highlights ⬇️