The volume of AI-enabled decent scientific work will overwhelm scientific journals and reviewers. Good post about this except it ends on a strange note, that Wiley and Elsevier AI tools might save us, which is both unlikely and would give even more power to them.
open.substack.com/pub/thebsdet...
Research is learning, publication is stage performance (well mostly)
Great work. Sadly applied econ had become pseudo-science thanks to causality obsession
English language has limitations to satisfy the identification gestapo. Any word somehow may imply causality.
They should announce a list of allowed words in descriptive works.
Applied Econ world is sometimes so Orwellian.
#stats #appliedecon
causation does not imply correlation.
A story of 21st century: The Syrian Esad continues his medical education in Moscow after a long career break when he ruled Syria for years including the civil war years. He will start his practice next year :)
No way to discuss. Fixed effects is religion for economists.
Fixed effects
I have been teaching undergrad econometrics for years. No ideal textbook exists, I am afraid.
The ideal should make clear distinction between description, casual inference, prediction.
I use Gelman for basic regression, The Effect for causality and Intro to statistical Learning for prediction.
Taking the literature as a whole, the global consequences of unmitigated climate change are likely to be substantial, unequal, harmful in aggregate, and potentially destabilizing, from Solomon Hsiang www.nber.org/papers/w34357
Great read! I wish I knew more about physics and chemistry, though.
Still point taken: stay away from beef.
Whatever you use, stay away from econometrics. Regression should be taught by statisticians.
This was the book that made regression enjoyable for me: sites.stat.columbia.edu/gelman/arm/ Regression and other stories is a modern version of that. But the first one was better.
all this happens because academia is no longer a medium of debate and discussion. It is mostly about individual performance and reputation.
We just need 1-year pause only to read others work
I really like @thomaspiketty.bsky.social’s paper Brahman Left, Merchant right.
Idea is simple yet profound: the political orientation of the rich has shifted from right to left after 1990s.
If you wanna be a good social scientist, learn how to make your adversary a caricature and then demolish it. Nobody else will read your opponents anyways.
That is 51 percent of high-level social science. If not more.
Bias-variance trade-off at work.
Even better, teach undergrads the difference between prediction and causal inference and description.
Highly relevant to anybody working with regression discontinuity designs
Looking forward to this…
Dear universities,
I am begging you to stop requiring letters of recommendation for master's programmes. You and I both know you don't read them, so stop asking for them.
Instead, have applicants list a name and get in touch if it's a borderline case.
Signed,
Everyone.
You break if you do not bend
All that led to a catastroph and they u-turned to high interest rates and monetary austerity. The result is overvalued lira, fall in real wages, highly uncertain climate. Yet: they still build roads.
This pretty much defines all right-wing movements of the last decades. Turkey’s Erdogan is exception though. He built lots of roads. Not much else than roads, though.
Economic history has benefited from convergence to applied economics through better use of data.
Conversely, I hope, it also helps applied economics to focus on important problems - not butterfly effects that represent 0.000001 standard deviation of the outcome of interest.
5. The Survivors - those who’ve figured out how to game the system just enough to get a good permanent job and never cares about anything any longer.
The list is not exhaustive.
4. Intellectuals - The beautifully obsessed ones, still chasing that one Big Idea that will revolutionize everything. Some get Nobel eventually.
3. Poseurs - Masters of academic kabuki theater. They know who to friend with, and who to co-author. lots of networking at the moment right conferences.
2. Teachers - the beloved professors students remember decades later and the ones who’ve been teaching the same notes since the 1990s.
My amateurish characterization of the personalities in academia:
1. Researchers - Those who can smell good article ideas from three departments away. Lots of projects and grants. Some max quality, many max quantity (topic does not matter).
Precisely. misplaced and too much emphasis on identification has created a crisis of relevance in applied economics, as well.
