Best of #econtwitter - Week of March 5, 2023: interesting tweets
Welcome readers old and new to this week’s edition of Best of Econtwitter. Please submit suggestions — very much including your own work! — over email or on Twitter @just_economics.
Neat illustration of the bias–variance tradeoff in analysis of a regression discontinuity... But then the variance turns into bias with file drawer bias vincentbagilet.github.io/causal_exagger…
Old-person observation: the establishment of six years as a normal length of an economics Ph.D. is bad, because behaviorally it makes it a lot easier to waste years 2 and 3 not trying to write a job market paper. Individually, people are better off taking 5 years as the goal.
Have JMPs gotten better as a result of students coming in with pre-doc experience and staying 1-2 years longer? My sense is that the top 5% of JMPs have gotten better but the median is essentially unchanged…?
I know a person who runs lots of field experiments. Literally prize winning stuff. Their estimate of how likely it is for a field experiment to happen successfully once a partner firm or organization agrees: 10% (Success here means actually carried out, correctly)
OK, made me make a version for econ bloggers... ❤️
I apologize for this meme https://t.co/N2wToRE7MB
Noah Smith 🐇🇺🇦 @Noahpinion
Can ChatGPT do a basic game theory problem set, with a focus on Nash equilibrium? Let's find out! Bear in mind that I have answered many of these questions (or cognates) online before (e.g., youtube.com/playlist?list=…), so this should be an easier case for ChatGPT than normal.
In the UK researchers asked a large group of people to keep a diary of what they did over a given day, and record how much they enjoyed each episode in the day. This chart shows the average enjoyment scores [From this piece I wrote a couple of yrs ago: ourworldindata.org/time-use-livin…]
Here is my favorite, a graph of various foods and their correlation with BMI. Lettuce: fatter. Dandelion greens: thinner. Olive oil: thin. Margarine: fat Fake sugar from plants: thin. Fake sugar from chemicals: fat This is because of bias in who eats what. Period. (2/2)
“The main reason why you give a seminar is to advertise the work. Originally, I thought the point was to get comments, but I now believe it’s the networking; people get to see the work, they get to understand it, and you get to clarify what they don’t understand.” Daron Acemoğlu
🚨🚨New Data Release🚨🚨 I'm satisfied enough to release my long-awaited dataset on municipal incorporations that does not rely on the Census of Governments or the Census Boundary & Annexation Survey. A 🧵 github.com/cbgoodman/muni…
Wow - amazing new data resource. Areas that can be reached by public transport in different timespans from all across the uk
While we haven't done anything on what services actually ran (& Tom's work on this is excellent), you can find loads on where you should in principle be able to reach (travel area) in the work we have published @DataSciCampus: https://t.co/OfLijo17zp https://t.co/vd3DW1Eja0
Arthur Turrell @arthurturrell
It’s the fancy way to say either “too fucking right” or else “get a load of this guy” or “listen up fuckos” as you prepare to rebut or further your argument. that’s why. https://t.co/ohSc0qfxsY
Why do academics love the word "indeed" so much???
Amit Goldenberg @Amit_Goldenb