|
| 摘要: |
| 本文研究了基于面板数据的分位数回归模型的变量选择问题.通过增加改进的自适应Lasso惩罚项，同时实现了固定效应面板数据的分位数回归和变量选择，得到了模型中参数的选择相合性和渐近正态性.随机模拟验证了该方法的有效性.推广了文献[14]的结论. |
| 关键词: 面板数据 分位数回归 自适应Lasso 变量选择 渐近正态性 |
| DOI： |
| 分类号:O211.7 |
| 基金项目:国家自然科学基金资助（11201356）. |
|
| PARAMETER ESTIMATION AND VARIABLE SELECTION IN THE QUANTILE REGRESSION MODEL FOR PANEL DATA |
|
HE Xiao-xia,XU Wei,LI Huan,WU Chuan-ju
|
| Abstract: |
| In this paper, we consider the variable selection problem for the quantile regression model based on panel data. By adding an improved adaptive lasso penalty term, we realize the quantile regression and variable selection for the panel data with fixed effect simultaneously, and obtain the consistency and asymptotical normality for the selection of the parameters. Simulation studies show the validity of the proposed method, which extend that of[14]. |
| Key words: panel data quantile regression adaptive lasso variable selection asymptotic normality |
