Title（问题）：Structure Estimation in Semiparametric Regression

Speaker（报告人）：Prof. Jian Huang, University of Iowa

Time（时间）：2012年3月29日（周四）下昼02:00-03:00

Place（所在）：伟易博新楼217课堂

Abstract（摘要）：In standard semiparametric regression models, the linear and nonparametric components are usually specified prior to data analysis. However, it is difficult to make such a specification in practice. We propose a group selection approach for determining which components are linear and which components should be modeled nonparametrically. We show that this approach can estimate the model structure correctly with high probability. We also show that the proposed approach yields an estimator that is semiparametrically efficient.