Last week, we hosted the First Chapel Hill - Copenhagen Conference on Macroeconomics and Deep Learning @UNC_econ. The conference was jointly organized by @JeppeDruedahl, @rabinovich_stan, Will Jungerman and myself. We were honored to have a brilliant keynote by @UncertainLars.
Today I'm presenting my new paper, “Macroeconomic Effects of Production Networks: A Deep Learning Approach,” at the SI 2026 Impulse and Propagation Mechanisms meeting, with Christian Matthes (@cmatthes_econ), Hikaru Saijo (@hpfilter), and Felipe Schwartzman.
The talk focuses on the substantive (and more interesting) findings about the dynamic properties of production and investment networks, but it builds on the Deep Sets architecture I developed with Mahdi Kahou (@MahdiKahou). That work leans on the symmetry group (a great unknown in economics, unlike in physics) and on the powerful implications of symmetry and permutation invariance.
There will also be a brief cameo of non-removable singularities in perturbation solutions, a point that, unfortunately, many macroeconomists do not fully appreciate.
If you are at the NBER SI, come by. It might be fun. And @FlorinBilbiie will present a really cool paper after me.
Under Smooth Diagnostic Expectations overreaction increases smoothly with uncertainty, providing a micro-foundation for key properties of survey data and the business cycle.
New paper by @Francesco_Bia, @CosminIlut, and @hpfilter
https://t.co/tWTAhjTx2H
#REStud#EconX
Smooth Diagnostic Expectations, joint work with @CosminIlut and @hpfilter, is out at ReStud. We build on the representativeness heuristic to show that overreaction is increasing in uncertainty. The paradigm can also account for overconfidence and other stylized facts of surveys
Under Smooth Diagnostic Expectations overreaction increases smoothly with uncertainty, providing a micro-foundation for key properties of survey data and the business cycle.
New paper by @Francesco_Bia, @CosminIlut, and @hpfilter
https://t.co/tWTAhjTx2H
#REStud#EconX
University of California STEM professors want standardized tests back due to severe math deficiencies among students:
“We now observe preparation gaps so severe that instructors must reteach middle school mathematics”
“The current admissions metric, based primarily on GPA & essays, can no longer reliably distinguish readiness for university-level STEM majors in an era of severe grade inflation & AI assisted application essays”
@cmatthes_econ This makes me pessimistic on two fronts. 1) Obviously, models are getting scary good and 2) Are labs going to be making most scientific progress while normal academics only have access to models that are say 6mo-1year behind the frontier??
Some used to give Dynare a bad rap for enabling DSGE-slop (not my view), but it was an important verification engine (and symbolic!) even in the pre-LLM era. Now that we’re speed running macro research with AI agents, I think it’ll become more and more important.