面板数据分位数回归模型的工具变量估计

Instrumental Variables Estimation of Quantile Regression Model with Panel Data

针对含有内生变量的面板数据回归模型,提出基于工具变量的分位数回归估计方法.首先,通过引入工具变量解决协变量的内生性问题,然后利用分位数回归的方法对回归系数进行估计.在一些正则条件下,证明所提出估计的大样本性质,通过模拟研究证实该方法的有限样本性质.

In this paper, we consider the panel data regression model with endogenous variables. Aquantile regression estimation is proposed based on the instrumental variables. To deal with the en- dogenous variables, we introduce some instrumental variables. The regression coe cient is estimated by the quantile regression. The large sample property is established for the proposed estimators under some regularity conditions. Simulation results illustrate the nite sample properties of the proposed method.