About

Hello! I am a PhD Candidate in the Department of Economics at the University of California, Davis. My primary research interest is econometrics, including both theory and applications. On the theoretical side, I focus on policy learning and partial identification of treatment effects. On the applied side, my work is closely connected to a range of policy-relevant questions in public, health, and development economics. I am on the 2025–2026 job market.

You can reach me at yuatuo@ucdavis.edu

Research Interests

Econometrics, Applied Econometrics

Working Papers

Optimal Treatment Under Spillover Effects and Noncompliance

(Job Market Paper)

Abstract. This paper considers optimal treatment allocation in the presence of spillover effects in both take-up decisions and outcomes. When there are two layers of spillover effects, I allow an individual's outcome to depend on other individuals' compliance types in the neighborhood. I construct estimators of the welfare function and the budget constraint, then based on the characteristics of the target population, I propose treatment allocations that maximize the estimated welfare subject to the estimated budget constraint, or minimize the estimated cost subject to a welfare target. I show that the proposed allocations achieve asymptotic welfare efficiency and weak feasibility. In an anti-conflict experiment in public middle schools in New Jersey, I estimate the minimum allocation required to achieve various targets for the average expected outcome. The results show that failing to account for both layers of spillover effects can lead to a suboptimal level of treatment allocation.

Causal Inference in RCTs with Multiple Treatment Arms

(draft available upon request)

Abstract. When there is more than one treatment group in a randomized controlled trial (RCT) with noncompliance, local average treatment effects (LATEs) are not comparable as each treatment has different compliers. I propose a partial-identification approach to compare two treatments in the presence of noncompliance. In particular, under extra dependence assumptions on the copula of unobservables, I derive pairwise sharp bounds on the average difference between the treatment effects for those who would comply with both treatments, as well as for other subpopulations. In an experiment evaluating the impacts of different depression treatments on economic outcomes in India, I provide bounds on the average difference between two depression treatments for always compliers.

Work in Progress

Inference for Marginal Treatment Effect (MTE) Estimator (with Takuya Ura)

Identification of Causal Effects in Network Experiments Using IVs

Teaching

CV

Here is my CV.