Awesome, well-deserved, and a great moment for the devo field! I’m proud of these friends and colleagues, and of course (especially) of my former student, the indomitable Esther Duflo
BREAKING NEWS:
The 2019 Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel has been awarded to Abhijit Banerjee, Esther Duflo and Michael Kremer “for their experimental approach to alleviating global poverty.”
#NobelPrize
Guido W. Imbens of
@Stanford
receives the Nobel Prize in
#Economic
Sciences for “methodological contributions to the analysis of causal relationships." Imbens shares the prize with David Card and Joshua D. Angrist.
Read more about Imbens:
#economics
Today we celebrated the mostly harmless work of Master Joshway
@metrics52
as the undergraduate and graduate labor economics students at MIT. May we continue his growing legacy.
#NobelPrize
#nobelprizeeconomics
And now, for those who've been holding back, go forth and master 'metrics in Python: (many thanks to Matheus Facure for an awesome resource including code, data, guidance, and, especially, a dose of timely MM and MHE-inspired humor!)
For decades, I've been a student of RCTs. Last year, I was privileged to be in one (). And now, in this cool paper, I'm thrilled to be ... the treatment!
Training policymakers in econometrics: Evidence from Pakistan
Another strong write-up (this one, in excellent English, crisp and engaging.) An Unexpected Result: How Nobelist Guido Imbens Helped Kick-Start the “Credibility Revolution” | Stanford Graduate School of Business
The econometric world lost a Grand Master this week.
was teacher, mentor, and friend to me and many others. Gary changed my personal path 30 years ago, when I was privileged to teach the first metrics class of my career with him.
Oh, and I've still never seen
@metrics52
and
@iamjohnoliver
in the same room. I think they may be the same person - and that is no more ridiculous than Oliver's assertion that there is only one Olsen twin.
Come on, I can't be the only one who thinks they look alike!
2/2
Third in our series of 'metrics chats with Isaiah Andrews - check it out! How Is Econometrics Changing? (Josh Angrist, Guido Imbens, Isaiah Andrews) via
@YouTube
Today was a special day at
@MITEcon
: our undergraduate development econ class celebrated Esther and Abhijit's
@NobelPrize
with them, including a group picture followed by selfies -- yes, economists can be cool!! 1/3
From Master Stevefu, on the credibility revolution and the Princeton Industrial Relations experience -
Natural experimenters: Nobel laureates David Card, Joshua Angrist, and Guido Imbens | VOX, CEPR Policy Portal
Just took the chance to transition from compiler to always-taker. Lucky the research team knows their way round an “IV” ! Thanks to the awesome mRNA-1273 investigators for the chance to be in The Trial of the Century
In case you were having 2nd thoughts about that colonoscopy your doc ordered: Instrumental variables methods reconcile intention-to-screen effects across pragmatic cancer screening trials | PNAS
Interview with Josh Angrist, 2021 Recipient of the Nobel Prize in Economics: Episode 7 of Mixtape: the Podcast. I interview Josh Angrist, winner of the 2021 Nobel Prize in economics, Ford professor of economics at MIT, and dir...
We're super excited to have
@metrics52
joining us for a new econometrics course! The first vids are in post-production and will drop early fall. Learn more here:
The econometrics battle of the century is here: the fearsome Josh Angrist vs the devious Selection Bias!
Who will win this clash of wits?
Watch the video to find out:
Just-identified IV is little biased unless endogeneity is egregious. And bias is halved by screening on sign(first-stage t-stat). Just in: M. Kolesar and I show that among pretests of the form t>c, setting c=0 minimizes bias, while c>0 can increase bias
I also regret the nber paywall. The nber raises a little money from sellling access. Alternatives are available, like page charges to well-funded authors. I encourage like-minded affiliates to make their feelings known to nber leadership.
Special thanks to the staff of
@theNASciences
and to
@SWEambUSA
and the incomparable diplomats of
@SwedeninUSA
for a wonderful, memorable Nobel day. You brought the spirit of Stockholm to DC!
I'm thrilled to see these two education path-breakers recognized. I've known Rick Hanushek and followed his scholarship for decades. I haven't always agreed with Rick's conclusions, but he's never failed to engage and challenge me!
Congratulations to our
#2021YidanPrize
Laureates Professor
@EricHanushek
at
@HooverInst
& Dr Rukmini Banerji at
@Pratham_India
! They’re both addressing a crucial piece of the education puzzle: improving quality of education & learning outcomes at scale.
Thanks to the awesome
@MRevUniversity
crew for helping me make Nobel video news and for pointers on how to act natural on camera (hope again triumphs over experience)
My answer to With respect to careers in data science and analytics, what advantage or unique value is there in graduate studies in econometrics when compared with statistics, engineering, or computer science?
Causality, experiments, and potential outcomes: dig into the numbers with Master Joshway 🧙♂️:
Complete instructional module here 👨💼👩💻:
(more to come)
Read 'em and .. weep no more! Unlock the mysteries of our sacred texts by watching MRU's latest blockbuster 'metrics vid...
How to Read Economics Research Papers: Randomized Controlled Trials (RCTs) via
@YouTube
Have you ever wanted to ask Josh Angrist, aka
@metrics52
a question? 🙋♀️ Now’s your chance!
While we're in Boston to film the Nobel Prize lecture 😎, we're doing a Q&A w/ the ‘metrics master himself – ask your q's in the replies ⬇️ by Thursday for a chance to be included!
Cogent discussion of micro metric evidence used well in development policy and elsewhere:
The Generalizability Puzzle (SSIR) via
@SSIReview
Savor the contrast with evidenceless rhetoric from frustrated graybeards:
"Peer quality isn't school quality."
@metrics52
talking about measuring school quality using value-added models at the MIT School Access and Quality Summit.
#mitschoolsummit
Great write-up from health economist Sherry Glied in JAMA: The Credibility Revolution in Economics and How It Has Changed Health Policy via
@JAMAHealthForum
part of
@JAMANetwork
Training policymakers in econometrics:
🔹increases their responsiveness to causal evidence
🔹increases their desire to run RCTs
Evidence from a field experiment with deputy ministers: by Sultan Mehmood
@mrsultan713
, Shaheen Naseer, and
@Daniel_L_Chen
Want to learn the latest in estimating school quality w/ quasi-experimental data? Or the various IV tips & tricks
@BlueprintMIT
has accumulated over the years?
Check out this new-and-improved handbook chapter with Josh (
@metrics52
) and Chris (
@c_r_walt
)!
@rwyeh
@NEJM
Fret not, my epi-friend. See Tbl S2: 80% of the sample has 15-day data; follow-up be balanced (434/538 vs 410/521). 7.6% in Remdawg group died, 13.4 in Placebo. SE on this awesome 5.8 pt diff is .021. Kaplan-Meier hazard is mostly hazardous, yo.
My answer to Bill Easterly has suggested academic economics journals excessively reward causal identification at the expense of researchers’ ability to address “big questions” relevant to policymakers. Do you (dis) agree?
alright JP, Im throwin it dahn, on yunz to pick it up: Josh A.😎 and Josh A.I.👾 will watch half a dozen empirical seminars remotely and tweet comments anonymously. Listeners vote on who's who. I always say my comments are entirely predictable) Lemme know when ur ready
@metrics52
Josh, I think you would like your doppelganger's performance. He goes by the name "Least Harmful" and can execute all kind of acrobatics; front-door, back-door, IV,... you name it. A true partner in credibility. How about helping us shaping his motion?
#Bookofwhy
Wouldn’t it be nice if econ talks were chill like social work, where all are called “Dr”, young defer to old, and none interrupt except in assent? Not for me. My most significant seminar experiences were challenging, uncomfortable, and unforgettable.
(Setting: Northwestern Econ job talk)
This made me smile--
Burt Weisbrod: "I'm sorry, I didn't meant to interrupt you mid-sentence"
@saskatchewin
: "Wow, this is such a polite department"
It's true! We sometimes ask tough questions, but we aren't rude 😀
Check out this promising and convincingly high-impact low-cost strategy to stem learning loss, reported in a super-concise Nature HB write-up (a format coming soon to the QJE et al., I hope):
Simple and Credible Value-Added Estimation Using Centralized School Assignment - It works: Risk-controlled OLS VAM, validated by lottery, coming soon to a school district near you!
Rainfall can be independent of potential outcomes and therefore the reduced form have a causal interpretation and yet not satisfy exclusion and thus the causal interpretation is very hard if not impossible to interpret. This is an example of how you need independence & exclusion
19th Century doc John Snow (the OG, not GoT!) gets a well-deserved moment in the spotlight:
Here's his story in 300 words and 1 remarkable table:
Differences-in-differences is born!
As part of this week's Nobel celebration we shot a new convo series with Josh Angrist, Guido Imbens, and Isaiah Andrews!
They cover productive research partnerships, the future of empirical economics, machine learning & more!
#EconTwitter
Learn more:
Check out my essay draft regarding an undergraduate design-based econometrics course using
#masteringmetrics
by
@metrics52
instead of a standard textbook
“Another approach”! Why do anything else? IV in this scenario identifies the effect of treatment on the treated, presumably the target in the first place.
Thrilled to release
@BlueprintMIT
busing research, using new causal tools to uncover exciting new findings: Still Worth the Trip? School Busing Effects in Boston and New York
Autor and Duggan (2003): "Steady state calculations augur a further 40% increase in the rate of DI receipt" ()
As it happened:
DI recipients/adults 25-64 was 3.51% in 2001, 5.20% in 2016.
Growth is 48%. Pretty good predicting. Bad for America.
There's a 14% earnings disparity between grads of public and private colleges? But does that mean private colleges CAUSE higher wages? Or might selection bias be lurking here?
Watch Josh Angrist attack this question (with weapons!) in the full video:
An awesome ‘metrics conversation with the incomparable Isaiah Andrews - check it out! Research Collaboration Do's and Don'ts (Josh Angrist, Guido Imbens, Isai... via
@YouTube
Can charter success be scaled? Yes indeed! Partly by reducing teacher effects. High-performing charters teach teachers how to teach consistently and well.
(1/3) MIT SEII is proud to announce big news – we are relaunching as Blueprint Labs! Our mission is to use data, economics, and analytic tools to uncover the consequences of policy decisions and improve society. Check out our new website at !