Time: Friday, March 24th 10:00-11:00

Venue: Tencent Conference 318-768-663

Topic: Jackknife Mean Group Estimator for Heterogeneous Dynamic Panels with Short T: An Application to Effects of Minimum Wages on Employment Levels. Panels with Short T: An Application to Effects of Minimum Wages on Employment

Speaker Introduction: YANG Liying , PhD in Economics, University of Southern California. Research direction: Econometrics.

Organizer: Bay Area International Business School, Beijing Normal University

Report Content:

This paper studies how to estimate and infer the average treatment effect in the heterogeneous dynamic panel data model for the panel data structure with large sample size but small observation period of cross-sectional units, and A new MG-JK estimation method is proposed by correcting the mean group estimator by Jackknife method. It is proved in this paper that when n and T both tend to infinity and n/T4 tends to zero, the first-order bias in the original MG estimation method can be eliminated, and the new MG-JK estimation method has the root-n consistency. In order to ensure the effectiveness of the Jackknife method on finite samples, this paper proposes a sufficient condition for the existence of r-order moments of the MG estimator for the ARX(1) panel data model. At the same time, this paper reviews the research on the impact of the minimum wage policy on the employment level, and explains from the perspective of econometric theory why there are differences between the different estimation results in the existing literature. The MG-JK estimation method can effectively solve the problems caused by the short-term dynamic effects of the outcome variable and the heterogeneity of the short-term treatment effects in different cross-sectional observation objects to the estimation of the average treatment effect. Compared with the two-way fixed effects estimator, the MG-JK estimation method does not need to assume that there are parallel trends in time between different parallel trends or the staggered treatment design. This paper applies the MG-JK estimation method to the panel data of American counties from 2002 to 2011 for analysis. The results show that the long-term impact of the real minimum wage on the overall employment level is close to zero, but there is a significant long-term negative impact on the youth employment level.