Fun question. Suppose E[T|X] is linear, for binary T. Then OLS estimates treatment effects via E[Y(T-E[T|X])] / E[(T-E[T|X])^2] = E[Y(T-E[T|X])] / E[Var(T|X)] Whereas IPW estimates E[Y(T-E[T|X]) / (E[T|X]*(1-E[T|X]))] = E[Y(T-E[T|X]) / Var(T|X)] (cont. in next tweet)

Exogeneous

.Dagens Medicin har interviewet Kristoffer Panduro Madsen og @t_kjaer (DaCHE) om nyt studie, som viser helbredseffekterne ved insulinpumpebehandling til type 1-diabetes dagensmedicin.dk/type-1-diabete…

#CSDID is getting a make-over!. For large projects it gas slow. Some times very. The bottle neck Data management in large samples. New version is more efficient! Works with Mata, using multiple "frames".

EXCITING NEWS: DaCHE is expanding! We are looking to hire three postdocs in health economics. The positions will run for a period of three (perhaps four) years. The date of employment is negotiable… sdu.dk/en/service/led…

We did it!!! Groundbreaking news: Tidepool Loop receives FDA Clearance! This is the first ever fully interoperable automated insulin dosing app, cleared by the FDA that began as a patient-led initiative. Find the inside scoop here: tidepool.org/tidepool-loop

Diff-in-diff is interesting because you start with *such* a simple implementation - OLS with an interaction term and a fairly grokkable parallel trends assumption. And it's great! Then you realize that breaking that basic case at *all* makes you have to change *everything*

Is Economics self-correcting? We have a new discussion paper, joint with N Fiala & @flneubauer. Quick answer is: No, rather not. We reviewed all replications published as comments in the AER. Plus, we surveyed the authors. A short thread 1/. bit.ly/3y0wMYp #EconTwitter

A reminder! Calling all health economists for great academic discussions in Odense, Denmark DaCHE #NHESG #healtheconomics #EconTwitter

Fantastisk god nyhed! 😄:Behandlingsrådets nye anbefalinger: Ja til sensorbaserede glukosemålere ⁦Steno Diabetes Center Aarhus⁩ ⁦⁦Diabetesforeningen⁩ behandlingsraadet.dk/nyheder/2023/b…

How quickly can you write an abstract? 🤔 #NHESG2023 DaCHE

Do you know why you should never merge m:m? well, check this out: friosavila.github.io/chatgpt/merge_… So if you think you need merge m:m, probably you need to joinby! #playingwithstata Stata

In Stata 18, you can now use Callaway and Sant'Anna (2021) as part of the Stata shipped software! The commands are hdidregress and xthdidregress. We will keep updating the csdid command, too, so you can use it with other Stata versions. stata.com/new-in-stata/h…

#EconTwitter I have created a new DiD Resources page on my website. It has links to all of my DiD teaching slides, coding exercises, and packages I've created. All this info was technically already online, but hopefully this makes it easier to find it! jonathandroth.com/did-resources/

🚨 DID Paper Thread 🚨 Are you confused about which new DID estimator to use? A bit more than a year ago, I wrote a short paper that gives a simple answer. I just completed a revision of that paper so here’s a 🧵 to celebrate! 1/N #econtwitter #differenceindifferences

This is the greatest randomized controlled trial joke ever told

Time to celebrate! “DID estimators of intertemporal effects” accepted at The Review of Economics and Statistics (REStat), and very very fast did_multiplegt_dyn Stata & R commands available from SSC & CRAN, thanks to Mélitine Malézieux Felix K. Diego Ciccia & Doulo! Celebratory thread! papers.ssrn.com/sol3/papers.cf…

👌

New study out with Kristoffer Panduro Madsen DaCHE: Time preferences⏳impacts health outcomes in people with T1 Diabetes using adv technology. Higher impatience = worse glycemic control, BMI, and kidney function. Time to rethink chronic disease management? 👇 doi.org/10.1002/hec.48…