统计与决策2024，Vol.40Issue(8)：5-10,6.DOI:10.13546/j.cnki.tjyjc.2024.08.001
两维异质性面板分位数模型的双惩罚回归方法
Double Penalty Regression Method for Two-dimensional Heterogeneous Panel Quantile Model
摘要
Abstract
This paper focuses on the panel quantile model with two-dimensional heterogeneity structure of coefficients,and based on SCAD penalty function and MCP penalty function,proposes a double penalty minimum weighted absolute deviation ob-jective function,and at the same time carries out parameter estimation and two-dimensional heterogeneity structure identification.Then,ADMM algorithm is used to solve the objective function,and BIC information criterion used to select the optimal adjustment parameters through grid search.Finally,based on Monte Carlo simulation results,the finite sample properties of the proposed method are verified,and the practical data are also used to verify the application effect.The results show that the proposed method can accurately identify the two-dimensional heterogeneous structure,and that the accuracy of parameter estimation of Post estima-tor is close to that of Oracle estimator.关键词
两维异质性/面板数据/分位数回归/双惩罚Key words
two-dimensional heterogeneity/panel data/quantile regression/double penalty分类
数理科学引用本文复制引用
任燕燕,李东霖,王文悦..两维异质性面板分位数模型的双惩罚回归方法[J].统计与决策,2024,40(8):5-10,6.基金项目
国家社会科学基金资助项目(21BTJ046) （21BTJ046）
山东省自然科学基金资助项目(ZR2020MG035) （ZR2020MG035）
山东大学人文社科重大项目(22RWZD16) （22RWZD16）
