蔡宗武教授学术报告：A New Semiparametric Quantile Panel Data Model: Theory and Applications
报告题目：A NewSemiparametric Quantile Panel Data Model: Theory and Applications
报告摘要：Motivatedfrom modeling the impact of foreign direct investment (FDI) to economic growth,in this paper, we propose a new semiparametric quantile panel data model withcorrelated random effects in which some of the coefficients are allowed todepend on some smooth economic variables while other coefficients remainconstant. A three-stage estimation procedure is proposed to estimate bothconstant and functional coefficients based on the integrated quasi-likelihoodapproach and their asymptotic properties are investigated. We show that theestimator of constant coefficients is root-N consistent and the estimator ofvarying coefficients converges in a nonparametric rate. A Monte Carlosimulation is conducted to examine the finite sample performance of theproposed estimators. Finally, the proposed semiparametric quantile panel datamodel is applied to estimating the impact of FDI on economic growth using thecross-country data from 1970 to 1999.