Research Interests:


  • Primary Field: High-Dimensional Econometrics, Applied Econometrics

  • Secondary Field: Applied Microeconomics, Machine Learning


Working Papers:










Work in Progress:


  • "Inference on Dynamic Discrete Choice Models: Hyperbolic Discounting versus Subjective Expectations" (with Yonghong An and Ruli Xiao)

  • "Test for Change in Persistence in the Dynamic Panel Model"


Research Experience:


  • Research Assistant for Professor Chihwa Kao, University of Connecticut (Fall 2017, Spring 2018)         











Abstract This paper proposes a biased-corrected FE estimator for dynamic panel models under a large sample size. The proposed bias-corrected estimator has several advantages compared to other dynamic panel estimators: first of all, it works for panel autoregressive coefficient in (-1,1]. Secondly, it is more efficient than the bias-corrected first-difference estimator. Lastly, unlike most existing dynamic panel estimators, the consistency of the proposed bias-corrected estimator does not depend on the stationarity of the initial condition. This paper further extends the model to include exogenous variables and heteroskedasticity. Based on the asymptotic distributions of the FE estimator under large n and T, the bias-corrected estimators for the models with exogenous variables and heteroskedasticity are proposed. According to the Monte Carlo simulations, the proposed bias-corrected estimator outperforms the GMM-type estimators in a large sample size.