统计学系系列讲座之333期

时 间:2019年4月9日(星期二)13:30-14:30

地 点:史带楼205室

主持人:张新生 教授 复旦大学管理学院统计学系

主 题:Statistical inference on high-dimensional generalized linear models:  a refined de-biased approach

主讲人:Professor Bin Nan Department of Statistics, University of California at Irvine

简 介:

Bin Nan 教授是美国统计学会(ASA)和国际数理统计学会(IMS)的Fellow、以及国际统计研究会(ISI)Elected Member。目前担任统计期刊Statistics in Biosciences 和 Lifetime Data Analysis的Associate Editor。在JASA,AOS,AOAS,Biometrika等国际期刊上发表论文超过100篇,他的研究兴趣主要集中在生存分析、高维大脑图像的数据分析、纵向数据的变点分析等研究领域。

 

摘 要:

In the existing literature, "de-biasing" or "de-sparsifying"  the L_1-norm penalized estimator represents a very important line of methods for drawing inference in high-dimensional linear models, and has been extended to generalized linear models (GLMs). However, we found that the de-sparsified approach in GLMs may not completely recover the bias or deliver reliable confidence intervals. In this work, we primarily consider the case of n > p with p diverging and provide an alternative modification to the original de-sparsified lasso, based on directly inverting the Hessian matrix, that further reduces bias and results in improved confidence interval coverage. Theoretical justification for drawing inference on linear combinations of the regression coefficients has been provided. Extensive simulations are conducted to show the improvement. This is a joint work with Lu Xia and Yi Li.

 

                      

 统计学系 

2019-4-2