题目：Distributed Variable Selection in Quantile Regression
主讲：冯兴东
时间：2017年11月13日 15:30-16:30
地点：明德主楼 1030会议室
摘要：
Recently, large scale datasets appear frequently due to the development of techniques. Distributed computation has attracted attentions from statistician. Since quantile regression has been an effective alternative to the classic mean regression in many fields. However, computationally efficient quantile regression estimates for large scale datasets are less developed. In this paper, we consider the penalized quantile regression estimate obtained via a new efficient ADMM algorithm that could be implemented in a distributed manner, which has been shown to be able to carry out variable selection for massive data.
简介：
冯兴东老师于2009年博士毕业于美国伊利诺伊大学香槟分校（University of Illinois at Urbana-Champaign）统计系，并于同年进入美国国家统计科学研究院（National Institute of Statistical Sciences）担任研究员（Research Associate），从事统计学方面的研究。现为上海财经大学统计与管理学院正教授。冯兴东教授的研究领域集中在分位数回归模型、数据矩阵降维、大数据统计计算等方面的研究。在国际权威学术期刊之上发表学术论文20余篇。