讲座题目：Inference for High Dimensional  Quantile Regression

主讲人：美国乔治华盛顿大学  王会霞教授 

讲座时间: 2017年12月25日（星期一）3:00-4:00pm 

讲座地点：综合楼601会议室

主讲人简介：王会霞博士，复旦大学统计系获得本科（1999年）及硕士（2002年）学位，从美国伊利诺伊大学获得博士学位（2006年），现任乔治华盛顿大学统计系教授，曾于2012年获得美国科学研究基金会Career  Award及国际数理统计协会Tweedie  Award，担任美国统计协会杂志和统计年鉴等顶尖统计杂志的副主编， 主持了多项美国国家自然科学基金的项目。

摘要：

In this talk, I will  discuss inference for high dimensional quantile regression, and present two new  tests based on maximum-type statistics to detect the presence of significant  predictors associated with the quantiles of a scalar response. The first test is  based on the t-statistic associated with the chosen most informative predictor  at the quantiles of interest. A resampling method is devised to calibrate this  test statistic, which has non-regular limiting behavior due to a weak  identifiability issue. The second test is based on the maximum of score-type  statistics. We show that for diverging dimensions, the test statistic converges  to the extreme value distribution of Type I under the null hypothesis under some  regularity conditions. For finite samples, we also introduce a convenient  multiplier bootstrap method to construct critical values. The proposed tests are  more flexible than existing methods based on mean regression, and have the added  advantage of being robust against outliers in the response.