CodeChella Madrid, a live workshop May 27-30 located in (wait for it) Madrid, Spain is now available for registration. It'll be on diff-in-diff, synthetic control and other things. Details below. Sponsored by Mixtape Sessions and CUNEF.
I follow this couple on TikTok and saw them on Cameo so I asked them to help me manage my students' expectations for my new prep on history of economic thought. Best $11 ever spent.
What perks did you enjoy as a kid bc of your parents jobs? My parents ran a computing company and they had a huge computer with a very primitive couple of games I’d play. They also has a coke machine on the honor system and I drank those all the time.
Many students starting out for the first time ask "Which stats software should I use? R or Stata?" The debates have gotten stale so I hired the couple on TikTok again to come in and give their two cents.
A senior colleague did me a favor and I told that person that I couldn’t repay them, and they told me “intergenerational favors aren’t supposed to be repaid, they’re supposed to be passed forward to the next generation.”
Around 6 years ago, my best friend told me his wife was pregnant bc a 99% accurate pregnancy test came back positive. Like any good friend would, rather than congratulating him, I texted him 3 pages of Bayes rule showing there was only a 92.7% chance she was pregnant.
#Friendship
Want to live longer? Try going to the opera. Researchers in Britain have found that people who reported going to a museum or concert even once a year lived longer than those who didn’t.
Dude this is freaking hilarious. 20% of papers in the gene literature are screwed up bc freaking excel converted the gene identifiers TO DATE. I think this was on twitter before but a student showed it to me again yesterday
This new article in economic inquiry coauthored by
@nickchk
is absolutely terrifying. Seven competent people were given two causal studies to replicate from scratch. No two people even report the same sample size and the standard deviation across estimates was 4x SE of original!
Brouwer’s fixed point theorem in a 60 second TikTok. Absolutely brilliant. He gets it all done illustrated with a cup of coffee in under a freaking minute. This is the bar.
Check this. Today the Mixtape is:
- No. 1 in probability and statistics (new release)
- No. 18 in probability and statistics (books)
- No. 3 in econometrics (best seller) ahead of Tim Hartford and Mostly Harmless
- No. 1 in econometrics (new releases)
Thanks everyone!
Causal Inference: The Mixtape is forthcoming from
@yalepress
for a physical delivery date of January 26th 2021. But the online version is available now. The online version is free and tries to enhance the experience of the paperback. Here's the site:
I don’t know if this is a controversial take, an obvious take or both. But I have a position on econometrics curriculum I’d like to verbalize. I believe causal inference should be taught before econometrics, not after, and here’s why. 1/n
This paper finds that THREE zip codes in the United States are responsible for over 50% of the citations in the top five economics journals (as opposed to 12% in other disciplines).
Look at that distribution of psychiatric beds by country. Now here’s an interesting couple of facts: we have the worlds highest incarceration rate but one of the worlds lowest psychiatric beds. I seriously doubt the problem is demand. Many mentally ill are in prison or homeless
So I have an Econ joke that makes me laugh but actually no one else. I say “I drink too much at the intensive margin, but not the extensive margin”. Meaning, I don’t drink very often but when I do, I drink way too much. Anyway I drank too much last night and shaved my head
Well, it's now official. I just signed my contract with
@yalepress
to publish my Mixtape. Super excited to be working with
@econeditor
who along with my wonderful agent
@LindsayEdgecom1
believed (and believes) in the project. So thank you to both of them.
Causal Inference: The Mixtape (Yale University Press) in technicolor. Release date January 26, 2021, for a measly thirty bucks. Think of it as causal inference for the price of two one month subscriptions to the family Spotify plans if that helps.
"Causal Inference: The Mixtape" (v1.8) is online. Consider this a beta version. Feel free to share it, light it on fire, use it for your class, whatever. It has 108k words, making it slightly shorter than MHE.
I can’t stop watching it. The key to a good TikTok voice over is the little things; like subtle perfections in the mannerisms. Yes I’ve thought about this a lot so sue me.
This spatial RDD on mask mandates is really well done. It's entitled "Mask Mandates Save Lives" by Hansen and Mano. Also a nice example of a spatial RDD tbh. Heterogenous effects varying by attitudes about mask wearing. Short paper. Compelling graphics.
I’ve started this tweet several times. One voice says it’s not people’s business and over sharing. Another voice wonders if never talking out loud about these sorts of things robs people from knowing and being known. So I’ll try again. I’ve decided to change my relation with econ
This paper in Econ inquiry by
@nickchk
was one of my favorite articles of the year. I know that bc I think about it a lot and have taught it twice. It’s very original, required a lot of organization and labor inputs and has a lot of lessons.
For those wondering how I'm doing, I'm doing great. My 18yo son is now living with me full time and it's been such a real joy. My 13yo daughter comes over constantly. And our middle girl too. I have new projects that make me happy, and I have students I'm proud of. I'm happy
Can’t even explain yet to my non-Econ friends just what a revolutionary force these three economists who just won the Nobel created. My book on causal inference has a blurb by Guido on the back and I sent one to Angrist with the inscription “thank you for inspiring a generation”
I’m going to delete twitter for a month. Need a break from here. Anyone wanting to talk, feel free to text or call. My cell is on my vita probably. Good luck to everyone this next hard month. I’m unusually heart broken about the world.
My paper, "Craigslist's Effect on Violence Against Women" (briefly renamed "Craigslist Reduced Violence Against Women but I lost that rebranding by a vote of 2-1) is under review and posted online. Some new stuff including better writing!
This new paper in economic history by Tianyi Wang (PhD Pitt; postdoc Copenhagen) is timely and timeless. He uses variation in a racist populist show caused by topological factors to estimate the effect of the show on voteshare, as well as increase pro-Nazi groups. 1/n
This R&R at QJE abt the impact of COVID on economic output, as well as subsequent policies, sounds like an empirical macro paper on recessions that we could see more of going forward. It’s by Cherry,
@michaelstepner
and coauthors. The level of detail — it’s unreal. Read abstract.
Congrats to Luca Braghieri, Ro'ee Levy and Alexey Makarin for their 2022 AER examining the effect of Facebook rollout on mental health. I discussed this great paper a long time ago, I think during Covid. Not surprised it hit high. Check out this bad ass DiD event study plot.
I told Brittany and Michael about that JEL paper finding that PhD students have above average struggles with depression, anxiety, loneliness and more, and they have a message to share for students out there struggling.
If there's not a pencil in your bag, you didn't need it!
Causal Inference: the Mixtape is coming out in December/January from
@yalepress
. You can preorder here. All your wildest dreams will come true if you buy it (causal claim)!
Retweeting. But here is what I think. Angrist and Krueger in their JEP on instrumental variables said the best instruments came from deep institutional knowledge. Meaning, it comes from being intimately familiar with some phenomena. 1/n
Can any economists or social scientists explain *how* they discover natural experiments?
It often seems to turn on knowing a super-obscure detail about some large-scale programme. Is there some short-cut hack for finding these details.
Installing R or python for me basically involves doing about ten things I find online, none of which I understand, and just praying at the end of it, it's installed.
People think I’m kidding. I want to be chief economist of Tinder. My entire career literally has been abt sex markets, internet and platforms. My dissertation was on gender imbalances and it’s effect on sexual behavior. I’ve been obsessed with the platform and matching stuff 1/
We are all set with the cover art; Here it is! I love the cassette tape which
@econeditor
suggested a long time ago. And honestly, it sort of maps back to my Yeezus cover which is nice. My son gave me a playlist with a similar cover art so it’s cool.
Okay, I just uploaded 9 videos from last week's
#CodeChella
2021 to my youtube page. They're numbered. I turned off the video accidentally for
@pedrohcgs
and
@ZhaoBean
so I had to upload that one from an old class sadly.
We got home after a week at my moms and the cats were nowhere to be found. I’d had someone checking on them and was watching them on the security camera but they weren’t here. Well the door to the HVAC was left open and they’d gotten lost in the walls. Landlady gave permission.
I won something! I haven’t won anything since a couple of poetry prizes in 1999. It’s a cash award too and I’m spending all of it on a virtual assistant.
This visualization of how local linear regression works never gets old. It was the first tweet I ever bookmarked which by revealed preference shows I’m a massive nerd.
I’m curious about what intellectual events changed your life? I’ve said it before but just so I can go first: Becker’s Nobel speech made me go to grad school, Roth and Sotomayor made me think in terms of two sided matching, and George Box made me value what models were and did.
Thinking about this new Becker book. When you think abt it, all of us are likely going to die with tons of resting and unfinished papers on our computers. I heard Mulligan or Murphy said when they read them that many were “very interesting”. So that’s kind of cool.
Willow: “do I get to keep the $500?”
Me: “What $500?”
Willow: “The $500 Trump is sending me.”
Me: “That’s for me & your mom for raising you.”
Willow: “no it’s for me. If you didnt have me, you wouldnt be getting $500.”
Me: “if I didnt have you, I’d have a lot more than $500”
Just FYI - people in labor, education, and macro say human capital all the time. It’s an important concept. Might as well be offended for saying money.
Why do people fresh out of school want to make a ton of money and build social capital that will increase permanent income and influence? Man. I’m going to have to think about that one.
This is an important story in its own right, but there's a meta-level that's easy to miss. Why do so many smart young do-gooders, like Pete Buttigieg, decide to go to McKinsey out of college? What's the appeal?
My son many years ago while listening to “Ignition” on the way to school: “Why does he need to have an after party? Why not just one big party?”
Me: “Rising marginal costs and capacity constraints.”
My son: “That’s your answer to everything.”
Have been told NUMEROUS times by students how powerfully important and meaningful
#econtwitter
is to their lives and careers. I suspect this is an important community for the profession even if it’s just a select group.
I wanted to share with people on Twitter the total breakdown of sales of my book, Causal Inference: The Mixtape, published by Yale University Press in January 2021.
I've had this lecture called "hidden curriculum" that I usually lead with in my workshops. It's the stuff we don't learn which is central to 1) empirical work and 2) being an academic researcher. I am adding to it this morning and wanted to check what I left out. 1/n
A nice way to introduce causal inference is to say “prediction is abt y hat. Causal inference is abt beta hat”. Learned that from a friend. It sort of gives this nice concept for students who will be using the same equations both ways.
Inaugural
#JHR_Threads
is about the cool study “The Labor Market Effects of a Refugee Wave” by Giovanni Peri (UC Davis) and Vasil Yasenov (Stanford) published in the Spring 2019 of the Journal of Human Resources. In the last tweet is a link to a Youtube interview with Yasenov
I won’t ever get over this tbh. Young MC said I changed how he thought about the difference between correlation and causality. Young MC. Bust a Move. Pinch me. This is really almost identical to a weird dream I’ve had before.
Professors of economics. What definition of economics do you give on day one? I usually use the lionel Robbins one about unlimited wants and limited resources with a second definition emphasizing allocations.
Causal inference fans, students and professors:
@andrewheiss
has posted lecture notes for a new class he’s teaching on causal inference. 30 pages, single space, table of contents, great pictures.
Well this is nice.
@Stata
just wrote me and said they’re carrying my book in the stata bookstore. I didn’t know I was allowed to have that on the bucket list.
I predict this will be in the top 10 front pages of the
@nytimes
since it first began. The importance of data visualization to making your point cannot be overstated.
You know, if I had had linear algebra, stats, real analysis and econometrics as an undergrad, my entire econometrics education at Georgia would’ve been a stream of insights one after another. I write the way I write now to help the kid lost in class bc I was him.
My daughter just got a job at a tex mex restaurant as a server and she’s super excited. I’m here and going to give her a big tip which is basically a dad dream of mine.
You know screw it. Anyone can attend the DD code-chella concert, faculty or student or anyone. It’s at 2500 and some faculty are already enrolled anyway so anyone is welcome.
First year PhD students in economics: first year is hard for many of us. You may be thinking about quitting. It’s your life, friend, but let me give you a few tips of advice. First, you should try to calculated expected utility from staying for four more years vs leaving. 1/n