Estimation and Inference for Latent Group Structures with Structural Instabilities

15:30-17:00, Friday, September 8, 2023

I-206, Boxue Building, DUFE

Dr. Wenxing Huang is an associate professor at the Antai College of Economics and Management at Shanghai Jiao Tong University. Her main research interests include high-dimensional panel models, non-stationary time series, dimensionality reduction estimation methods, and their applications in financial econometrics and macroeconomics. Her papers have been published in journals such as Journal of Econometrics, Econometric Theory, Journal of Business & Economic Statistics.

In panel data models, the latent group patterns of cross-sectional heterogeneity may change over time. Treating the group structures as unknown, we propose a novel penalized least squares approach to consistently estimate the dates of structural breaks. Splitting the sample based on the estimated break dates, we then show that the Classifier-Lasso estimation of the group-specific coefficients, the group membership, and the number of groups remains consistent. Furthermore, Lasso-type approaches are developed to consistently identify the individuals affected by the breaks as well as changes in the group-specific coefficients. The proposed methods perform well in Monte Carlo simulations. In the empirical application, we revisit the classical Fama-MacBeth regression and find time-varying latent group structures in risk premiums.

For more information of the seminar, scan the following QR code(s) to join Tencent QQ group (904 544 292) or WeChat group named "IAER Seminar (4)", please.

QQ Group

WeChat Group 

(QR code is valid until 12/9/2023)