Research

Publications

Average Treatment Effect Localization: Projection Methods in Synthetic Control

Accepted in Econometric Theory. arXiv

Abstract. Many real-world policies and business interventions require assessing short-term effects to inform timely decisions, even though most causal inference methods focus on long-term average treatment effects. In this paper, we introduce average treatment effect localization (ATEL), which captures localized, short-term policy impacts in panel data settings with a single treated unit and provides early indicators of policy impact. To accommodate both time-varying and nonlinear effects of observed and unobserved covariates, we propose a nonparametric model for untreated outcome, interpreted as a time-varying factor model via sieve approximation. Estimating the time-varying factor model is challenging due to the boundary bias and identification. Our estimation method based on diversified projection can effectively address these issues. We develop an asymptotic distribution theory to facilitate inference for the ATEL estimator. In an empirical application, we apply our proposed methodology to assess the impact of right-to-carry laws on violent crime rate.

Working Papers

Overfitting Inference with Instrumental Variables

Latest Draft

Abstract. Overfitting is traditionally viewed as a problem to be mitigated in empirical economics. This paper shows that it can instead be a blessing for inference in high-dimensional instrumental variables (IV) models. We revisit the factor IV framework, in which the instruments and the endogenous variable share common latent factors, and show that the first-stage signal is dense, providing a theoretical justification for recent empirical evidence against sparsity. Moreover, the implied first-stage coefficients shrink toward zero as the number of instruments diverges, exhibiting “local-to-zero” behavior commonly used to describe settings with many weak instruments. Building on this characterization, we propose an IV estimator based on ridgeless regression combined with sample splitting. This procedure remains well defined when the number of first-stage regressors exceeds the sample size. We derive the asymptotic distribution of the estimator under regimes with (very) many weak instruments, and show that, when the number of instruments is sufficiently large, its asymptotic variance coincides with that of an oracle factor IV estimator that directly observes the latent factors. Monte Carlo simulations indicate favorable finite-sample performance relative to existing high-dimensional IV estimators. Empirically, we apply the proposed procedure to estimating the elasticity of substitution between immigrants and natives in the United States.

Work in Progress

Overfitting at the Cutoff: Inference in Regression Discontinuity Designs