Pedro H. C. Sant'Anna @pedrohcgs Twitter profile

Pinned Tweet

Pedro H. C. Sant'Anna

4 months

Are you using DiD to learn about something *continuous*? If so, then new DiD tools for binary designs don’t seem to help much. But here is something that will help! An update to “DiD with a Continuous Treatment” w/ @agoodmanbacon and Brant Callaway:

15

241

1K

Last Seen Profiles

@senn_heath

@PNGTourism

@LarryLoftis

@99pm_e

@davidkayevo

@dalydonofrio

@PTFCBR

@elfanek

@JimScalpert

@tarundsnaol

@AP_McCoy

@davidmoraless96

@haechandt

@thezcsd

@sailor_movie

@zachlew20

@blinkmf

@tianamelville25

@susannetaylor60

@emc_clark

@hatocube

@Nathan_Erbach

@hajikano_hiroki

@samanwaya_sr

@Yeah_mal

@kymshu_alacarte

@sonzaru

@siena_chas45931

@profitabill

@TheZetaPlan

@SpringfieldCece

@Laura_de_CUT

@_eatmysneaks

@stevepoltz

@ENikaloas

@el_e_l_l_e

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

🦶🦶🚨New DiD Paper🚨🦶🦶 What if treatment is continuous not binary? What parameters can you estimate under what assumptions? What about two-way fixed effects regressions? What about staggered timing? Brant Callaway, @agoodmanbacon and I dive in here:

13

327

1K

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

😍I'm coming back to Academia😍 I am joining @EmoryEconomics in September! My time as a full-time tech economist has been excellent. It has helped me become a better communicator and practitioner. But I am so happy and excited about this new chapter! Full-time researcher

114

16

1K

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

🚨Hello #EconTwitter ! I am very happy that my paper with Brantly Callaway, "Difference-in-Differences with multiple time periods", is now forthcoming at the Journal of Econometrics. What are the main take aways? I will ask my daughter to help me out. 1/n

26

260

1K

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

If you want to learn Causal Inference from one of the best and most productive researchers on the area, I have some good news for you!! @pengding00 lecture notes are here 👇👇👇

A First Course in Causal Inference

I developed the lecture notes based on my ``Causal Inference'' course at the University of California Berkeley over the past seven years. Since half of the students were undergraduates, my lecture...

arxiv.org

8

275

1K

Pedro H. C. Sant'Anna

@pedrohcgs

4 years

For all people doing IV in their work, this brand new paper seems *very* relevant! One important takeaway msg is: Rule of 1st stage F stats > 10 can be very misleading. Authors also give nice alternative inference tools that can “rescue” us all!

22

280

1K

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Hey #EconTwitter , #HealthPolicy , #PoliticalScience and #SocialScience in general. Do you feel lost about recent advances in DiD? We are here to help! What's Trending in Difference-in-Differences? A Synthesis of the Recent Econometrics Literature 1/n

9

262

1K

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Do you use Difference-in-Differences methods? Have you ever wondered how selection into treatment relates to parallel trends? The role of unobservables? @DaliaAGhanem , @wuthrich_k , and I are very excited to share our paper: 1/n

2

211

1K

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

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.

12

159

954

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

This is definetely the best feeling ever!!!Olivia Sant’Anna, dada loves you. Dada will also make sure you appear in his DiD #EconTwitter posts soon!!

77

4

875

Pedro H. C. Sant'Anna

@pedrohcgs

4 years

Time for another #Econometrics thread! Today I want to talk about my paper with Jun Zhao (absolutely great PhD candidate from Vanderbilt ), "Doubly robust difference-in-differences estimators", which is now forthcoming at the Journal of Econometrics! 1/n

19

165

756

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

I'm so happy that my paper with Brantly Callaway won the Dennis J. Aigner Award at the @JEconometrics . In fact, this is an award for the recent DiD advances, as CS is sharing the prize with 2 other DiD papers that y'all know and love: @agoodmanbacon and Sun and Abraham! Thanks!

58

748

Pedro H. C. Sant'Anna

@pedrohcgs

7 months

I am happy that my family is now part of the population that qualifies for being part of twinning IV *and* same-sex IV studies! Yes, we are expecting twins! And yes, I am now a Z=1!

48

3

701

Pedro H. C. Sant'Anna

@pedrohcgs

3 months

So, for y'all who've been asking for code, you will need to wait a bit =P Happy to announce that the twins, Gloria and Pedro, were born and are doing great! So is Mom! We are back home, and our older kids, Sofia and Olivia, are in love with their little brother and sister!😍😍

66

5

676

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

📈I am so excited to share with you my brand-new Difference-in-Differences course! This time, we are doing everything on our own! No third parties! Kind of cool, right?!!! Check our website: 1/n

9

80

600

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

📢📢The @Stata native implementation of Sant'Anna and Zhao DRDID, and Callaway and Sant'Anna DiD procedures are now available for everybody to use it.📢📢 @friosavila is a Stata wizard and made it v. easy to install: ssc install drdid, all replace ssc install csdid, all replace

10

144

599

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

I failed Micro I, and also the Micro and Metrics prelims the first time I took them… Some of these exams required some type of memorization, and I am really bad at that! Plus, I was not that focused on only studying With extra effort and maturity(?), things worked out!

Gaurav Sabnis

@gauravsabnis

1 year

Flex your lowest academic low score. To remind us that almost everyone fails at some point in life but it isn't the end of the road. Just a memory in the end. Mine was 27/100 in Engineering Math 2, just a year after I'd scored 99 in 12th boards. Much needed kick in my butt.

4K

2K

23K

3

26

576

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

Some professional news: very excited to be joining @Microsoft team in Seattle in a month from now. I will spend at least the next academic year there (on leave from @VanderbiltEcon ) I am really looking forward to put all the causal inference tools into practice.

41

8

573

Pedro H. C. Sant'Anna

@pedrohcgs

10 months

We've been talking about toxicity in Econ for a while now! And we very often ask ourselves: How can we improve? @GilDoVigor situation below tells that not everybody is on the same page about that. UC Davis lost 8 pages of his micro prelim, and the econ dpt is making him retake

GIL DO VIGOR

@GilDoVigor

10 months

Eu pensei muito se iria expor essa situação desta maneira. Mas eu sou uma pessoa que detesta injustiça e estou passando exatamente por isso, agora. Eu fui aprovado em todas as disciplinas do PhD, exceto a de microeconomia. @ucdavis @UCDavisGrad @UCDavisEcon

365

525

18K

2

27

534

Pedro H. C. Sant'Anna

@pedrohcgs

5 months

I am really happy to join the Editorial Board of the Journal of Econometrics ( @JEconometrics ) as Associate Editor. JoE is the home of many great papers! Plus, it's where I have published many of my papers! Please send your papers on #CausalInference and #DiD to us!

40

13

525

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

I’ve been thinking about what else I can do to make my Econonetrics research broadly available to empirical researchers, especially for the younger generation. So, if you are teaching any of my papers in your PhD course, and want me to talk to the students, let me know!

15

49

502

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

New working paper by @paulgp @instrumenthull and Michal Kolesar This seems very interesting and relevant!

5

125

500

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Came back from #ASSA2023 very motivated and grateful! Here's a picture of @agoodmanbacon , Liyang Sun, Brant Callaway, and me, just after we received the @JEconometrics Dennis J Aigner Award for the best empirical paper (our 3 DiD papers shared the award) Thanks a lot!!!

24

22

465

Pedro H. C. Sant'Anna

@pedrohcgs

8 months

I am now settled in Atlanta! These two helped a lot with the move!

9

0

439

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

Hey #EconTwitter ! I've partnered up with @StatHorizons to offer a short course on Difference-in-Differences. The details can be found here:

7

82

429

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

This was a lot of fun! We managed to get a lot of material covered in 4.5h: - DiD basics - functional form - staggered adoption - problems and solutions We answered 145 questions, live! All the materials are posted at The recording is coming soon!

UM Population Studies Center

@UM_PSC

1 year

Join the next @UM_PSC PDHP half-day workshop Jan. 31 on “Difference-in-Differences” analysis, presented by @pedrohcgs . Free and open to researchers & data analysts of all skill levels-- with hands-on practice using R and Stata. RSVP: . @umisr

6

75

247

15

66

421

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Who said that matching is a bad strategy?!!

12

3

419

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

I am so happy to see this out in the open! We are finalizing an R package to implement the test in more general cases, including staggered setups. So stay tuned!

Econometrica

@ecmaEditors

1 year

This paper by @jondr44 and @pedrohcgs shows that parallel trends holds for all functional forms of the outcome (eg levels, logs, percentiles) if and only if you have "PT of distributions". They also show this is testable and characterize when it can arise

1

73

311

14

57

420

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Twitter is fun because we can promote the work of friends and people we admire! Today I want to call your attention to a brand new paper by @jondr44 and @jiafengkevinc , available at: What is the gist? Check the abstract!

6

78

391

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

So good to see another recent and influential DiD paper finally published in one of my favorite journals: Congrats @agoodmanbacon !!

4

51

391

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

🦸‍♀️🦸Time to push some #DiD updates to #EconTwitter and the entire community who is enthusiastic about Difference-in-Difference methods🦸‍♀️🦸 We have been listening to all of you, and I am happy to report that did #R package is now SUBSTANTIALLY faster.

6

44

378

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

I am soo excited to see the December issue of the Journal of Econometrics @JEconometrics !!! There are 3 DiD papers y’all know: Sun and Abraham, Callaway and Sant’Anna, and @agoodmanbacon Bacon-decomposition.

3

36

369

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Many people ask me about DDD, and our DiD course at @CausalSolutions do cover it. Slides are here: The video recording of the lecture is now available on Youtube, too: Think of this as a sample of my DiD course =)

Triple Differences Research Designs at Causal Solutions

Difference-in-Differences (DiD) is one of the most popular methods in the social sciences for estimating causal effects in non-experimental settings. Its pr...

www.youtube.com

Jessica H. Brown

@JHB_econ

2 years

Hey #EconTwitter ! I'm still catching up on the diff-in-diff lit... Is it possible to do triple differences with the new methods?

16

24

197

2

70

365

Pedro H. C. Sant'Anna

@pedrohcgs

2 months

One month as a family of 6!! These 4 kids make everything worth it!

10

0

361

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Seeing my DiD paper with Brant, Callaway and Sant’Anna (2021), being used at Microsoft and Xbox is very very cool!!!

8

4

351

Pedro H. C. Sant'Anna

@pedrohcgs

3 months

I'm impressed and grateful for the care that the @EmoryEconomics community has been sharing with my family since the twins were born. We got flowers from the department, the grad students cooked food for us, several faculty sent gifts… I feel seen and loved. Thanks for that!

4

351

Pedro H. C. Sant'Anna

@pedrohcgs

5 months

Merry Christmas from The Sant’Anna family! The gift of love. The gift of peace. The gift of happiness. The gift of hope. The gift of prosperity. May all these be yours at Christmas. May God bless you and your loved ones throughout the year.

12

2

336

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

Dada, given the recent trends, I think I am ready to start my pre-doc/RAship so I can be a competitive candidate when/if I apply for a PhD in Econ

2

7

327

Pedro H. C. Sant'Anna

@pedrohcgs

4 years

Here I go, on another #Econometrics related thread. Today, I want to talk about the "debate" related to health policies, economic growth and the 1918 Spanish flu. Everything I have to say is here (with codes): Let's get to it! 1/n

3

80

318

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

🚨Updated Working Paper Alert🚨 "Efficient Estimation for Staggered Rollout Designs", joint with @jondr44 (). This paper is all about how we can do better than DiD in setups with quasi-random treatment timing! 1/n

4

64

310

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

This starts with all of us teaching how to do code review and data management during the PhD! This is probably the biggest and most impactful push we, economists, should be getting from Data Science. - Post your codes on @github - Document it - Have a code reviewer, even if

4

35

301

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

About a 1.5 years ago I posted the paper by Lee et al on weak IV and the F-test>100. That got many people attention, and that was probably a good thing! Well, many things happened since then, and their paper is now conditionally accepted at AER! 1/n

Pedro H. C. Sant'Anna

@pedrohcgs

4 years

For all people doing IV in their work, this brand new paper seems *very* relevant! One important takeaway msg is: Rule of 1st stage F stats > 10 can be very misleading. Authors also give nice alternative inference tools that can “rescue” us all!

22

280

1K

2

63

296

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Today was essentially my last day at Microsoft. I’m finishing this chapter happy. I learned a ton! I had the pleasure to spend my last day talking with Bing about Callaway and Sant’Anna (2021), as they are using it in several project! And so is Xbox! Really glad to have helped!

5

3

293

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

Sofia and I took the afternoon to catch up on our #stats #econometrics readings... She is hungry for knowledge, as you can see... She is taking all #EconTwitter advices seriously and working on that pre-doc really early on!

6

290

Pedro H. C. Sant'Anna

@pedrohcgs

10 months

The 2023 Meeting of the Brazilian Econometric Society will be in Rio! We will have: - Jeff Wooldridge @jmwooldridge for a mini-course - Guido Imbens @guido_imbens as Keynote - Martha Bailey @martha_j_bailey as Keynote + more It'll be awesome! I hope to see you there!

Sociedade Brasileira de Econometria (SBE)

@SBEconometria

1 year

🗣️Atenção!! Abertura de submissão de artigos para o 45º Encontro da Sociedade Brasileira de Econometria. O Encontro será Windsor Barra Hotel, no Rio de Janeiro, no período de 12 a 15/12. 📝Prazo de Submissão: 01/06 a 27/07/2023 🔗Link de submissão:

4

25

107

7

33

287

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

📢📢Hello #EconTwitter , #PoliSciTwitter , and all people curious (or overwhelmed) about the advances in Difference-in-Differences methods! In partnership with @StatHorizons , I will offer a new session of my DiD course on Oct. 14-16, 2021. Information:

2

64

259

Pedro H. C. Sant'Anna

@pedrohcgs

19 days

The videos of how the police mishandled the situation at Emory yesterday are very troubling. Prof. Caroline Fohlin is a fierce defender of all Emory students; anyone who knows her knows that. It was also clear in the videos! It is just awful how things turned out.

11

21

255

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

I cant get tired of seeing these. You can clearly see the joy and happines in Guido’s and also @Susan_Athey ’s faces. This is soo inspirational.

Stanford University

@Stanford

3 years

"He really likes to make sure he understands things. He'll keep pushing until ... the math just sings to the ideas and the ideas sing to the math." - @Susan_Athey on her husband and co-author Guido Imbens, recipient of the 2021 #NobelPrize in economics

7

87

465

1

21

250

Pedro H. C. Sant'Anna

@pedrohcgs

16 days

Let’s talk about the worst publishing experience. I will go first: I had a paper a Econometric Theory that was conditionally accepted after 3 rounds of review. Something bizarre happened, Editor sent it back to referees. We addressed the points, but it didn't matter: rejected!

31

5

233

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Hey #EconTwitter and otherwise! On December 15th, we will have a webinar about recent advances in DiD methods. Speakers are @lihua_lei_stat , @CdeChaisemartin , @jondr44 and #twitterless Brantly Callaway. More info: Will be fun! See y'all there.

Andrej Srakar

@SrakarAndrej

2 years

Wednesday, December 15th, One🌎YoungStatS presents 𝗔𝗱𝘃𝗮𝗻𝗰𝗲𝘀 𝗶𝗻 𝗗𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝗰𝗲-𝗶𝗻-𝗗𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝗰𝗲𝘀 𝗶𝗻 𝗘𝗰𝗼𝗻𝗼𝗺𝗲𝘁𝗿𝗶𝗰𝘀. 9:00 PT/12:00 EST/18:00 CET. Register👉👉 @YoungStatS2 @pedrohcgs @StatFen @BernoulliSoc @InstMathStat

1

23

67

4

61

228

Pedro H. C. Sant'Anna

@pedrohcgs

8 months

Tomorrow, September 21, at 2pm EST, Amit Gandhi (Airbnb) and I will chat about the usage of different Causal Inference tools in the tech industry as part of the CausalAI conference! Everything will be live on Youtube: Many other cool talks, too!

0

46

225

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

#EconTwitter What is your favorite DiD applied paper? Here, I must add that the data must be publicly available so I can actually replicate/run the files. I want to use something other than minimum wage this time!

13

39

221

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

I get asked a lot about how to do DiD with Poisson regression. The key is the type of Parallel trends assumption you want to impose. A brief overview on 2x2: These are the types of questions we can answer in our DiD course at

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

@MayraPinedaT It'll depend on your Parallel Trends assumption! With additive PT, you can do it (did package does not support it, but you can modify). But with Poisson, you probably want a "ratio-in-ratio" estimator (relative PT) I've written a bit on this here:

2

40

4

32

215

Pedro H. C. Sant'Anna

@pedrohcgs

4 years

Today I want to give a shout-out to @TymonSloczynski paper, "Interpreting OLS Estimands When Treatment Effects Are Heterogeneous: Smaller Groups Get Larger Weights", that is currently available here: Paper is conditionally accepted at ReStat!

5

51

216

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Spent last few days remotely at my father’s farm/home, where internet speed is 1Mb/s and electricity arrived less than 10 years ago. But you can taste the glavor of real fruits. Plus, you gain a lot of perspective. Rural Brazil has its perks and I really needed this!

6

1

213

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

Friday afternoon R&R!!

7

0

214

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

For everybody asking on how to implement Callaway and Sant’Anna DiD (csdid) in Stata, here you go!!! Trial version is out. If you try it,keep us posted about it.

Fernando Rios-Avila (~ ‾⌣‾)~

@friosavila

3 years

and Just a bit after the two-week mark, #CSDID is ready for testing! This version estimates the asymptotic version of all estimations and aggregations. But yes, the WB version will be coming! For details check it out here: (1/6)

6

21

92

2

43

210

Pedro H. C. Sant'Anna

@pedrohcgs

4 months

TWFE linear regressions are the best tools for DiD designs. This story of negative and non-transparent weights never matter in practice, so people can ignore the nee DiD methods. Plus, if the weights are convex, we can settled as they answer clear causal questions. 🙈🙈🙈

❈ 𝘒𝘈𝘌 ❈

@HoeFeelin

4 months

QUOTE THIS WITH YOUR RESPONSE 👀

51

10

184

4

7

208

Pedro H. C. Sant'Anna

@pedrohcgs

10 months

People make mistakes. We are humans. However, we must own things up if we care about student success and well-being and really do not want to perpetuate toxicity as the status quo. If the faculty lost 8 pages of @GilDoVigor prelim, he's not the one who should incur this cost!

1

3

206

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

Doubly robust DiD Stata package on its way!!!

Asjad Naqvi

@AsjadNaqvi

3 years

#drdid update on #Stata implementation: Its done. I will do some more testing and convert it into some usable form. Doing a full package will take time but I will put online in a week or so. @pedrohcgs @causalinf @nickchk

6

21

138

3

17

201

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

I'm so excited that a 2nd cohort of our 𝘿𝙞𝙛𝙛𝙚𝙧𝙚𝙣𝙘𝙚-𝙞𝙣-𝘿𝙞𝙛𝙛𝙚𝙧𝙚𝙣𝙘𝙚𝙨 𝙘𝙤𝙪𝙧𝙨𝙚 at @CausalSolutions is coming up in March 2023. In total, 30+ hours of DiD! Detail info: #EconTwitter #CausalTwitter #PoliSciTwitter

3

41

203

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Olivia’s first birthday!!❤️❤️❤️

6

1

201

Pedro H. C. Sant'Anna

@pedrohcgs

4 months

This is a fair point to be made, and Jon’s writing is clear as usual! Brant and I have discussed this, and that is why we have the “universal” baseline option in our did R package. Brant also has a post related to this: Easy to address!

Universal vs. Varying Base Period in Event Studies

Introduction

bcallaway11.github.io

Jonathan Roth

@jondr44

4 months

#EconTwitter Have you replaced your dynamic TWFE event-study with an event-study from one of the recent DiD methods for staggered timing? If so, you may be interested in this short note on interpreting event-studies from these new methods. A short 🧵

13

312

1K

3

29

197

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

The Zoom link with the recording is out:

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

This was a lot of fun! We managed to get a lot of material covered in 4.5h: - DiD basics - functional form - staggered adoption - problems and solutions We answered 145 questions, live! All the materials are posted at The recording is coming soon!

15

66

421

1

39

195

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

I have mentioned several times that one of my favorite recent DiD papers is the @asheshrambachan and @jondr44 "A More Credible Approach to Parallel Trends" (forthcoming at Restud, available at ). They've just released a @Stata implementation, so go use it!

Jonathan Roth

@jondr44

2 years

🚨 HonestDiD is Now Available for Stata 🚨 Check it out: Major kudos to Mauricio Cáceres Bravo, who made the Stata package faster and easier to use than the R one

14

120

634

0

47

190

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

During my PhD, I often complained I was spending so much time TAing and taking classes far from my field. Fast-forward 10 years. I am now using the material I prepped as TA for a “Economics of Network Industries” course in my work at MSFT! Who would have guessed?!!

4

2

191

Pedro H. C. Sant'Anna

@pedrohcgs

4 years

For the next month or so, I plan to mention an Econometrics paper that I really enjoy reading. I will try to do this every Monday. The first paper of this series is by @asheshrambachan and @jondr44 , "An Honest Approach to Parallel Trends" Link here:

2

34

191

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Let’s say it louder so people in the back can hear us well!

6

8

188

Pedro H. C. Sant'Anna

@pedrohcgs

7 months

🔥This Friday, @wuthrich_k will present Selection and Parallel Trends, our joint work with @DaliaAGhanem , at the Chamberlain Seminar🔥 In this updated version of the paper, we also discuss how to conduct selection-aware sensitivity analysis for DiD. 📰:

1

39

188

Pedro H. C. Sant'Anna

@pedrohcgs

4 months

🚨 It is still early in the Job Market season, but we are happy to announce that Florian Gunsilius (from U Michigan) will join @EmoryEconomics in Fall 2024. Welcome to the team, Florian! And, TBC, we are not done: we have another Econometrics position to fill! #EconTwitter

8

11

189

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

“It was not long ago that all I talked about was TWFE <> DiD. Now, all I care is giraffes, zebras, and the entire zoo!” #happybirthday #twoyearsold #SlowDown

5

1

185

Pedro H. C. Sant'Anna

@pedrohcgs

10 months

I cannot imagine what this type of issue would have done to me and my mental health if this happened to me back in the day. I am really hoping that UC David finds a fast resolution for this...

1

2

185

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Merry Christmas!!!!🎄🎄🎄🎅🏼🎅🏼🎅🏼🎄🎄🎄

7

0

182

Pedro H. C. Sant'Anna

@pedrohcgs

2 months

"I enjoyed reading your paper that makes a useful contribution to the literature on DiD models, [...]. Unfortunately, I have concluded that since your contribution is primarily a methodological one, it is not an ideal fit for a general-interest journal like the AER."

໊

@buffys

2 months

what’s an insult you’ll never forget? 😭

227

70

1K

7

6

180

Pedro H. C. Sant'Anna

@pedrohcgs

9 months

These two make me so proud every single day!

6

1

181

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Every day that passes, I am realizing that I like applied microeconomics with solid econometrics more and more!! Yes, I think I really am an applied econometrician 😱😱 Even though I still like talking and learning about semiparametric efficiency now and then!

3

179

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Spent last couple of days figuring out how to reliably do Triple Differences with covariates. Setups 2 periods, setups with multiple time periods, and also with staggereed treatment timing. All unified Very cool stuff, especially when you want a Doubly Robust estimator!

6

5

177

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

If you are interested in a 14h long course on DiD, check it out 👇👇👇👇

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

📢📢After the big success of the sold-out DiD course earlier this summer, @StatHorizons and I have decided to offer another session on 24-27 August, 2021.📢📢 Information: Will cover a lot of material! Join us! #EconTwitter #RStats #causaltwitter

2

37

154

1

26

170

Pedro H. C. Sant'Anna

@pedrohcgs

5 years

Here is something I do not quite follow: Why do people almost always refer to two-way fixed effects (TWFE) models as synonymous to Difference-in-Differences (DID)?? TWFE is an *estimation method*, while DID is a *research design*. These are very different things!

9

27

170

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Update: back home, recovering well! Doctor removed it with success. Thanks everyone for the supporting and caring messages!!!

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Kidney stones are a bitch!

11

0

71

15

0

169

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

It was just great!! @causalinf #CodeChella 2021 DiD seminar was impressive! 6-7 hours of DiD for everybody. I also loved the chat feature of Twitch! It was great to chat with @jmwooldridge , @taylor_wright , @friosavila and so many others.

7

6

169

Pedro H. C. Sant'Anna

@pedrohcgs

1 month

One million times this!!!!!!! It is all about making the comparisons you want, those that respect your identification assumptions!

Peter Hull

@instrumenthull

1 month

Regression is a tool for making comparisons If you don't know / can't easily explain what comparisons you're trying to make, then you don't understand the regression you're running

2

100

621

0

9

167

Pedro H. C. Sant'Anna

@pedrohcgs

6 months

Are you applying for an Economics Ph.D.? Are your students thinking about it? Consider our growing program @EmoryEconomics ! GRE is optional & you can apply for a fee waiver here Check out our placement & join us!

5

34

166

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

Excited about the #ASSA2023 in NOLA! I'll present an update of the "DiD with Continuous Treatment" paper with Brant and @agoodmanbacon . What is really cool is that the new estimators are showing exciting results in the application! Want an early peek? Come to our session!

4

21

165

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

Card + Angrist + Imbens!!!

2

6

166

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

“Why did you settle for a non-interpretable causal parameter?!”

7

1

156

Pedro H. C. Sant'Anna

@pedrohcgs

3 months

Today I experienced what true peace looks like: All four kids asleep for two hours, at the same time, uninterrupted! God, thanks for this!

8

0

159

Pedro H. C. Sant'Anna

@pedrohcgs

1 month

I am excited about some new DiD-related projects I have been working on! It has been a long time since I felt this excited tbh! 2 more weeks and the teaching semester is over. Then, I enter that magical period in that I believe I can do everything (but not?) Finger crossed!!

4

2

158

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Hey #EconTwitter , just want to let y'all know that Microsoft's Office of the Chief Economist is hiring researchers@ This is a sweet position where you can keep doing research but also get involved with product work. It's very cool seeing things in action!

7

29

157

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

It feels so good to be back in Nashville after one year in Seattle! Even better, I can continue pushing my econometrics and causal inference research and keep engaging with practical and exciting real-life problems! Working at Microsoft has been really cool!

3

0

156

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

📢📢After the big success of the sold-out DiD course earlier this summer, @StatHorizons and I have decided to offer another session on 24-27 August, 2021.📢📢 Information: Will cover a lot of material! Join us! #EconTwitter #RStats #causaltwitter

2

37

154

Pedro H. C. Sant'Anna

@pedrohcgs

10 months

It is very cool to see my @EmoryEconomics webpage up: Looking forward to interacting with the entire @EmoryUniversity and the broader Atlanta #Economics and #CausalInference community!

0

6

156

Pedro H. C. Sant'Anna

@pedrohcgs

4 years

This is a very nice overview of the different (regression-based) "Machine Learning" techniques that are now becoming popular in econometrics/economics.

1

23

152

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

In DiD with 2 groups (treated and untreated), and short T, should we cluster at the "treatment" level and give up inference? Shouldn't the choice of wether to cluster depend on the type of sampling/counterfactual you are thinking? I really need help! @jmwooldridge @arindube

7

18

155

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

The joy of seeing this is soo great!!! Congratulations!!!!

The Nobel Prize

@NobelPrize

3 years

BREAKING NEWS: The 2021 Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel has been awarded with one half to David Card and the other half jointly to Joshua D. Angrist and Guido W. Imbens. #NobelPrize

194

6K

12K

1

3

153

Pedro H. C. Sant'Anna

@pedrohcgs

1 year

It's been 2 weeks since I started working as an Amazon Visiting Academic within the Supply Chain Optimization Technologies (SCOT) org. It's been amazing. The team's knowledge and engagement with causal inference tools are just 🤯 Shoutout to @amazon @AmazonScience So much

2

6

152

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

I think most people don't realize how much work it is to maintain and keep adding new features to a package. I honestly love doing this, but it is very demanding.

scott cunningham

@causalinf

3 years

Software for empirical work is such a double edged sword. On the one hand, if it’s easier to do, then under uncertainty, you’ll use the easier one. So software can create dependencies and probably arms races therefore to get that software made first. 1/n

7

4

52

5

10

152

Pedro H. C. Sant'Anna

@pedrohcgs

2 years

Created this place so we can have a centralized place to share, discuss and ask questions about Difference-in-Differences and related econometrics methods. Feel free to join #EconTwitter

2

40

149

Pedro H. C. Sant'Anna

@pedrohcgs

3 years

Potential outcomes meeting DAGs in Seattle - @analisereal #EconTwitter

6

4

147

Pedro H. C. Sant'Anna

@pedrohcgs

2 months

I am glad to see @ReceitaFederal willing to share data for research. But I want to know what is the process to having access to such data, too! Many other Brazilians that I know are eager on leveraging this info for better policy making, including the great @ursulammello

Gabriel Zucman

@gabriel_zucman

2 months

Good bye São Paulo! So grateful for the warm hospitality I received for my first visit to Brasil And now back to work — we have a report to write on the minimum tax on the super-rich, and a great research collaboration starting with @ReceitaFederal Can't wait to be back

5

26

329

3

13

147