My research interests are in theoretical and applied econometrics. I work on identification problems for more credible causal inference, as well as nonparametric estimation and statistical inference for interesting policy questions.

Working Papers

• Inference for an Algorithmic Fairness-Accuracy Frontier (2025), with Francesca Molinari
  Extended abstract in Proceedings of the 25th ACM Conference on Economics and Computation (EC’24)
  [arXiv] | [code] | [EC'24]
• Using Forests in Multivariate Regression Discontinuity Designs (2025), with Alice Yuan Qi
  [arXiv] | [code]

Work in Progress

• Synthetic Parallel Trends (Job Market Paper)
  

Abstract: I show that popular empirical strategies for policy evaluation in the panel data literature—including difference-in-differences, synthetic control methods, and their variants—all rely on key identifying assumptions that can be expressed through a specific choice of weights $\omega$ relating pre-treatment trends to the counterfactual outcome. While each choice of $\omega$ may be defensible in empirical contexts that motivate a particular method, it relies on fundamentally untestable and often fragile assumptions. I develop an identification framework that allows flexible choice of weights by introducing a new assumption, Synthetic Parallel Trends: the treated unit’s trend is parallel to a weighted combination of control units’ trends for a general class of weights. The framework nests these existing methods as special cases and is by construction robust to violations of their respective assumptions. I characterize the identified set for the treatment effect of the treated unit, which admits a linear programming representation that yields moment equalities. I provide a consistent estimator and a valid confidence set for this identified set, and illustrate their finite-sample performance in a simulation and empirical value in an application.