I'm not quite sure how to articulate this, but I feel we need to do a better job of teaching econometrics students how to ask coherent questions. This post is not mainly prompted by the advent of AI, as I've been thinking about it since I started spending time on statalist.

Parenting choices, shaped by economic incentives, contribute to social stratification and inequality in society, from Francesco Agostinelli, Matthias Doepke, Giuseppe Sorrenti, and @FabrizioZilibo1 nber.org/papers/w33963

Couldn’t have done it without the incredible support from my committee and the entire SMU Economics. Thank you all for guiding me through this journey and making SMU a place of growth and inspiration. Onward to the next chapter!🎓💙❤️ #PhDone #EconTwitter

Neural networks (e.g., LLMs) make often imperfect predictions, introducing biases into analyses that rely on them. Common empirical economics scenarios fall outside the existing literature on debiasing “black-box AI”. Our paper (with Jake Carlson) on robust and efficient

Bottom line: if you want your structural estimates to be more robust to model misspecification, be sure to use recentered instruments! arxiv.org/abs/2504.04056

We introduce a novel experimental design and econometric model to estimate risk aversion, prudence, and temperance without imposing their interdependence. Our approach unmasks variation in risk attitudes that existing modeling approaches may obscure. econometricsociety.org/publications/e…

I'm happy to report I sent the final draft of my "TWFE-Two-Way Mundlak" diff-in-diffs paper to Empirical Economics, to appear in a special issue in honor of Robin Sickles. I took out some of the material in the original working paper but added more practically useful stuff.

Who becomes an inventor? Three main factors dictate patenting rates: income, intelligence, and exposure to innovation. People are greatly influenced by the environment in which they are brought up, and that really matters. Let’s take a comprehensive look! 1/

Estimation and Inference in High-Dimensional Panel Data Models with Interactive Fixed Effects. arxiv.org/abs/2206.12152

Quasi-Bayes in Latent Variable Models. arxiv.org/abs/2311.06831

You can just be an econometrician.

Instead of testing for bias from potential misspecification of model restrictions, it's nearly optimal to adapt, averaging restricted and unrestricted estimates using data-driven weights. shinyApp: lsun20.github.io/MissAdapt/ Paper: econometricsociety.org/publications/e…

"A new method for identifying and estimating consumer preferences when choice sets are constrained due to selection or search, but are not observed" New paper by Agarwal & Somaini: restud.com/demand-analysi… #econtwitter #REStud

Let me explain why I believe modern economics is such a powerful tool for understanding the world. I’ll do this by discussing a great paper by Simone Cerreia-Vioglio, Lars Peter Hansen, Fabio Maccheroni, and Massimo Marinacci, “Making Decisions Under Model Misspecification,”

Cross-Validated Causal Inference: a Modern Method to Combine Experimental and Observational Data. arxiv.org/abs/2511.00727

Unbiased Regression-Adjusted Estimation of Average Treatment Effects in Randomized Controlled Trials. arxiv.org/abs/2511.03236

Paul G-P (paulgp), Michal Kolesár, and I have a new guide to leniency designs (aka judge/examiner IVs), prepared for the JEP The bottom line? Keep calm and UJIVE on!

Glad this paper with guido imbens is out in the JEP. The LaLonde paper has had a big impact on my academic journey and continues to teach us about the challenges and possibilities of conducting credible inference using nonexperimental data. aeaweb.org/articles?id=10… Many thanks

Optimally-Transported Generalized Method of Moments. arxiv.org/abs/2511.05712