题目：Shrinkage Quantile Regression Estimation for Panel Data Models with Multiple Structural Breaks

报告人：张立文 讲师（上海大学）

地点：瑞安楼609室

时间：2017年12月13日（星期三）15:30-16:30

报告摘要

In this paper, we consider a high-dimensional quantile regression model for panel data with multiple structural breaks. We develop a penalzied estimation method with both slope coefficients and individual fixed effects by combining the informations over multiple quantile levels. We show that with probability tending to one our proposed method can correctly determine the unknown number of the the breaks and estimate the common break dates consistently. The asymptotic distributions of the Lasso estimators of the regression coefficients and the post Lasso versions are also established.

Simulation results demostrate that the proposed method works well in the finite samples. The perfomance of the the proposed method is futher illustrated by the analysis of a environmental Kuznets curves data.

个人简介

张立文，讲师，上海大学经济与管理学院老师，毕业于复旦大学统计系，师从著名统计学家朱仲义教授，主持国家自然科学基金青年项目，发表多篇高水平SCI期刊论文。