当前位置:首页 > 学术交流 打印页面】【关闭
3月21日加州大学圣地亚哥分校周文心博士讲座预告
( 来源:   发布日期:2017-03-20 阅读:次)

 

讲座题目A New Perspective on Robust Regression: Finite Sample Theory and Applications

讲 座 人:University of California, San Diego (加州大学圣地亚哥分校)

周文心博士  Assistant Professor

讲座时间:20170321日(周二)下午15:00---16:00

讲座地点:浙江工商大学行政楼统计学院601会议室

主讲人简介

       周文心,2013 年获得香港科技大学统计学博士学位(导师:邵启满教授),随后在澳大利亚墨尔本大学数学与统计学院进行博士后研究(指导导师: Aurore Deleigle)。主要研究方向为:asymptotic theory in probability and statistics, large-scale statistical inference, nonparametric and robust statistics.近几年,有多篇文章在概率统计学顶级学术期刊The Annals of Probability, The Annals of StatisticsBiometrics等杂志发表。

Abstract:

       Massive data are often contaminated by outliers and heavy-tailed errors. To address this challenge, we propose the adaptive Huber regression for robust estimation and inference. The key observation is that the robustification parameter should adapt to sample size, dimension and moments for optimal tradeoff between bias and robustness. Our framework is able to handle heavy-tailed data with bounded $(1+\delta)$-th moment for any $\delta>0$. We establish a sharp phase transition for robust estimation of regression parameters in both finite dimensional and high dimensional settings: when $\delta \geq 1$, the estimator achieves sub-Gaussian rate of convergence without sub-Gaussian assumptions, while only a slower rate is available in the regime $0<\delta <1$ and the transition is smooth and optimal. As a consequence, the nonasymptotic Bahadur representation for finite-sample inference can only be derived when the second moment exists. Numerical experiments lend further support to our obtained theories.

欢迎老师和同学们参与讨论!


上一条: 3月29日统计学学术沙龙“高水平期刊学术论文写作经验交流” 2017-03-28
下一条: 3月24日北京师范大学邱东教授讲座预告 2017-03-20