A related issue is that, if you ask people on surveys, they often respond that it is more valuable to extend life for older/sicker people with low remaining life expectancy. Is that rational? More generally, there are some open theoretical questions about how to aggregate in these situations.
That said, this seems like a good idea for the world we’re in today. Authors who want their work to be widely read should have incentives for this.
I’m surprised by how pessimistic this is about PDF parsing. I would expect most of these extraction issues to be largely solved within the next 5-10 years.
llms.txt is a good idea, although there too I wonder how much value added it will have in 5-10 years.
Seems like a real milestone: researchers have integrated an emulated fly brain into a simulation. The video is spooky
theinnermostloop.substack.com/p/the-first-...
The abstract submission deadline for HHES is coming up in 10 day (March 15). It's always a great set of papers and an amazing setting, so don't miss it!
tinyurl.com/HHES2026-Inv...
Between personal machines and the central HPC cluster there’s also the middle layer: department or lab servers. Feels like those might see renewed demand too.
Harsh Parikh will present work on how to transport estimated effects from one set of networks to another.
Shuangning Li will share new results on covariate adjustment in experiments.
So if you're working on networks+causality, consider participating. Abstracts due March 10th causnets.github.io
The Criminal Justice team at Arnold Ventures is looking for a pending/recent PhD with strong causal inference skills, for a remote, part-time consulting position.
This is a great opportunity for someone considering a transition from academia to the policy space.
Deadline: March 15
Please share!
Great episode! One small note: at the beginning it sounded like you said humans have about one trillion heartbeats in a lifetime; it's actually on the order of ~3 billion
DeclareDesign is super useful for anyone in the design state of their experiment
Medicare Part B spends over $40 billion on physician-administered drugs. Our new study, forthcoming in the Journal of Public Economics, finds that the program's payment formula slows drug price growth over time.
Are you a PhD Student curious about working in topics of health and aging? Do you feel like you may need extra support? I have a program for you!
A one-day mentoring workshop hosted by yours truly and Jetson Leder-Luis (BU) through @nber.org
www.nber.org/calls-papers...
and also: tikz!
#EconSky: Registration for "The Craft of Publishing in Economics" is now open! March 9, 12-1:30 EDT.
Three fantastic editor will give suggestions for how to navigate the publication process w/ plenty of time for Q&A.
More info & registration form below.
econ4ua.org/the-craft-of...
⚙️ Working at the intersection of causality and networks?
We're organizing a satellite event at @netsciconf.bsky.social in Boston on June 1st. The focus is networks science and causal inference.
Submit your work by March 10th!
causnets.github.io
It probably got overloaded by a surge of eager health researchers!
📢 Predoc Opportunity
I'm hiring a full-time predoc to work with me at UC Berkeley, supporting my research on labor economics, the economics of education, and public policy.
jesse-rothstein.com/rothstein_pr...
Please pass on to anyone who might be interested.
#econsky #econ_ra
New dataset dropped this month: national Medicaid provider spending (2018–2024), aggregated by NPI x HCPCS x month. Includes FFS, managed care, and CHIP encounter data. Public, 3.3GB, provider-level detail. Looks very useful for utilization & spending research
opendata.hhs.gov
AI may end up being part of the solution, making it easier to quickly read and assess papers. For editors facing a potential flood of submissions (I haven't seen it yet, but it may be coming...), AI might raise screening capacity as fast as it raises volume.
#EconSky: Save the date & tell your friends! Economists for Ukraine will host an online workshop called "The Craft of Publishing in Economics" on March 9, 12-1:30 US EST.
Sign up here to get more info:
docs.google.com/forms/d/e/1F...
The Francesca Gino saga has reached its logical conclusion: the LinkedIn post about what my adversity has taught me to launch “consulting services”.
Good idea
#EconSky: Tymofiy Mylovanov from the Kyiv School of Economics is a simply a hero.
While Trump is busy allying with Putin, Tymofiy is out in the cold working for his students and his country.
I will match all donations to KSE, up to $1K total (link below).
www.youtube.com/shorts/RgDXI...
Nice essay by Carlos Chavez on the history and development of causal inference
carloschavezp29.substack.com/p/on-causality
Pb concentration by decade in hair from Salt Lake City region residents. Value plotted for 1940 includes all samples from 1916 to 1959; value plotted at 2022 includes all samples from 2020 to 2024.
Lead (Pb) in archived hair in Salt Lake City.
"Lead (Pb) in archived hair documents a decline in lead exposure to humans since the establishment of the US Environmental Protection Agency "
www.pnas.org/doi/10.1073/...
"Among patients already admitted to the hospital when a ransomware attack begins, in-hospital mortality increases by 34–38 percent."
From the thread:
> Surprisingly, ChatGPT was much better at explaining and helping with Claude Code than Claude itself. For example, Claude gave me incorrect info about how to set up its permissions files.
I still see a good amount of Econ content here, and have decent engagement on my posts, but yeah, it’s not (yet?) the same as the old days of EconTwitter
Agree, my use of git will be increasing going forward.
I'm also wary about blanket write permissions. For now, I gave it only write permissions in the github directory ("Write(**)" in settings.json)
Will let you know how it goes with data analysis!
