Best of Econtwitter

Share this post

Best of #econtwitter - Week of January 30, 2022 [4/3] special metrics edition

www.bestofecontwitter.com

Best of #econtwitter - Week of January 30, 2022 [4/3] special metrics edition

An Economist
Feb 2, 2022
Share this post

Best of #econtwitter - Week of January 30, 2022 [4/3] special metrics edition

www.bestofecontwitter.com
Share

Because there were already three parts to last week’s newsletter (part one, two, three), broken out here is a high concentration of good metrics content from last week:

Twitter avatar for @theJohnBonney
John Bonney @theJohnBonney
Hi there #econtwitter—you may have noticed my new paper with @cblandhol, Magne, and Alex: "When is TSLS Actually LATE?" I am stoked about this paper and wanted to briefly summarize our findings. (1/n) Paper here: papers.ssrn.com/sol3/papers.cf…
Image
5:37 PM ∙ Jan 28, 2022
249Likes75Retweets
Twitter avatar for @theJohnBonney
John Bonney @theJohnBonney
Our findings? Without additional parametric assumptions, there is precisely one 2SLS specification that is *guaranteed* to be a positively-weighted average of LATEs. Remember that SW specification? The one that no one ever uses? Yep, that’s the only one. Oof. (9/n)
5:38 PM ∙ Jan 28, 2022
18Likes2Retweets
Twitter avatar for @theJohnBonney
John Bonney @theJohnBonney
Where does this leave us? Well, unless you are using the SW approach (and we both know you aren’t), then under the usual IV assumptions alone, 2SLS is *not* LATE. You must also assume that you have rich covariates and that your 1st stage is monotonicity-correct. (13/n)
5:38 PM ∙ Jan 28, 2022
Twitter avatar for @matt_blackwell
Matt Blackwell @matt_blackwell
Kudos to everyone who is figuring out each case, but maybe just make your working hypothesis that if you’re using least squares with anything other than a fully saturated discrete covariates, you may not be getting what you expect.
11:24 PM ∙ Feb 1, 2022
Twitter avatar for @agoodmanbacon
Average Weighter @agoodmanbacon
Every day I stray farther from the view that controlling for things using a regression is a good idea.
7:42 PM ∙ Feb 1, 2022
241Likes16Retweets

---

Twitter avatar for @instrumenthull
Peter Hull @instrumenthull
Ok, so I come bearing good news for ~93% of you: esp. those bootstraping complex models (e.g. w/many FEs) Instead of resampling, which can be seen as reweighting by a random integer W that may be zero, you can reweight by a random non-zero non-integer W
Twitter avatar for @instrumenthull
Peter Hull @instrumenthull
Question for economists who do empirical work: When you need to bootstrap your SEs, do you:
4:54 PM ∙ Jan 29, 2022
479Likes96Retweets

^thread, lots of discussion in the replies/QTs, truly the best of econtwitter, including:

Twitter avatar for @grant_mcdermott
Grant McDermott @grant_mcdermott
Okay, here’s my quick attempt at making this a fast, reusable function. gist.github.com/grantmcdermott… I _think_ it’s working, but please kick the tyres (examples at the bottom). If all looks good then I might port to a package.
Twitter avatar for @instrumenthull
Peter Hull @instrumenthull
Ok, so I come bearing good news for ~93% of you: esp. those bootstraping complex models (e.g. w/many FEs) Instead of resampling, which can be seen as reweighting by a random integer W that may be zero, you can reweight by a random non-zero non-integer W https://t.co/Rpm1GmomHg
3:44 PM ∙ Jan 30, 2022
56Likes6Retweets
Twitter avatar for @jiafengchen42
Kevin Chen @jiafengchen42
Seems like today is a good day to talk about why the bootstrap works Here's my intuition---
Image
9:05 PM ∙ Jan 29, 2022
411Likes38Retweets
Twitter avatar for @paulgp
Paul Goldsmith-Pinkham @paulgp
Here's a nice write-up by Bryan Graham on this method here: bryangraham.github.io/econometrics/d… also Chamberlain and Imbens (2000): nber.org/papers/t0200 (ungated older version here: scholar.harvard.edu/chamberlain/fi…)
Image
Twitter avatar for @instrumenthull
Peter Hull @instrumenthull
Ok, so I come bearing good news for ~93% of you: esp. those bootstraping complex models (e.g. w/many FEs) Instead of resampling, which can be seen as reweighting by a random integer W that may be zero, you can reweight by a random non-zero non-integer W https://t.co/Rpm1GmomHg
8:48 PM ∙ Jan 30, 2022
23Likes5Retweets

---

Twitter avatar for @jiafengchen42
Kevin Chen @jiafengchen42
Excited to share a new preprint with @DRitzwoller on semiparametric estimation (read, *machine learning* estimation) of long-term treatment effects, such as those identified by a latent unconfoundedness or a surrogacy assumption (1/n) arxiv.org/abs/2107.14405
2:03 AM ∙ Jan 31, 2022
69Likes17Retweets
Twitter avatar for @causalinf
scott cunningham @causalinf
Couple things I'm learning today: 1) Limited information maximum likelihood (LIML) originates in the Cowles Commission work on simultaneous equation modeling and instrumental variable methods in the 1940s and 50s by Haavelmo, Koopmans and others 1/n
1:41 PM ∙ Jan 29, 2022
73Likes11Retweets
Twitter avatar for @economictricks
Francis DiTraglia @economictricks
I'm coming around to the idea that we need to teach DAGs. The textbook advice is to control for anything that's correlated with both treatment and outcome without being caused by the treatment. This is wrong, but it's very hard to explain why verbally. @causalinf #EconTwitter
2:19 PM ∙ Jan 28, 2022
140Likes18Retweets
Twitter avatar for @rmkubinec
Robert Kubinec @rmkubinec
Econometrics is this really confusing mix of "your standard errors could be off asymptotically, use this non-parametric clustering adjustment algorithm" and "it's fine if the data are binary, just use OLS, no one cares about the errors anyway." #polisci #econtwitter
4:37 AM ∙ Jan 26, 2022
499Likes37Retweets
Share this post

Best of #econtwitter - Week of January 30, 2022 [4/3] special metrics edition

www.bestofecontwitter.com
Share
Comments
Top
New
Community

No posts

Ready for more?

© 2023 An Economist
Privacy ∙ Terms ∙ Collection notice
Start WritingGet the app
Substack is the home for great writing