题目:High-dimensional subgroup functional quantile regression with panel and dependent data
汇报人: 梁汉营
会议时间:2025年12月5日(周五) 15:30-17:00
地点:综合楼644会议室
报告人简介:梁汉营,同济大学数学科学学院教授,博士生导师。现为中国现场统计研究会高维数据统计分会常务理事,中国现场统计研究会大数据统计分会常务理事。1997年博士毕业于武汉大学,1997-1999年在中国科技大学作博士后研究。研究兴趣:不完全数据的统计分析,分位数回归,高维数据分析,贝叶斯分析,经验似然,变点分析。主持过国家自然科学基金面上项目5项、国际合作项目1项和教育部项目2项,发表学术论文140余篇,曾获第十一届全国统计科研优秀成果奖二等奖、重庆市自然科学二等奖以及安徽省自然科学三等奖。
报告摘要:In this talk, we focus on high-dimensional additive functional partial linear single-index quantile regression with high-dimensional parameter under subgroup panel data. Based on spline-based approach, we construct oracle estimators of the unknown parameter and functions, and discuss their consistency with rates and asymptotic normality under a-mixing assumptions. A penalized estimation method by using the SCAD technique is introduced to estimate the additive functions and parameter, enabling variable selection. Hypothesis testing for the parameter is also considered, and the asymptotic distributions of the restricted estimators and the test statistic are derived under both the null and local alternative hypotheses. Simulation studies and real data analysis are conducted to verify the validity of the proposed methods and applications.
