Economic and Social Development Institute Seminar (No.71)

Topic: A panel data probit model with correlated random effects and time varying factor loads

Date: Nov. 10th, 2017 

Time: 1:30 p.m.-2:45 p.m. 

Venue: Meeting room 106B, Zhonghui Building

Speaker: Zhou Ya-hong, Shanghai University of Finance and Economics

Sponsor: Economic and Social Development Institute, JNU

ABOUT ZHOU YA-HONG

He is now the professor and doctoral supervisor of School of Economics in Shanghai University of Finance and Economics. He is New Century Excellent Talent, and got doctoral degree in Hong Kong University of Science and Technology in 2005. His research field covers Microeconometrics and Microeconomics. His research achievements were published in Econometric Theory, Journal of Business & Economic Statistics, Science China Mathematics, Journal of Econometrics, Economics Letters, Annals of Economics and Finance, Economic Research and other famous journals home and abroad. 

ABSTRACT

We study inference in a panel data binary response model with correlated random effects and time varying factor loads. The model is useful in that it allows returns on unobserved individual effect (labor skill) to change over time, e.g., in a female labor force participation model. By assuming that the composite error term is Gaussian distributed, we show that both the slope coefficients as well as the factor loads are identified up to scale. We propose a multi-stage estimator that is shown to be root-n-consistent and asymptotically normally distributed. The proposed estimator has a closed form expression and thus is very easy to compute. Simulations and the empirical illustration demonstrate the usefulness of our estimator.