学术科研

您现在的位置: 首页- 学术科研- 学术动态

【预告】统计学院系列学术报告(2016年第17期)

报告时间:20161229日(周四)上午10:00

报告地点:京师学堂第七会议室

报告人简介: Liqun Wang(王力群),Professor and Head of the Department of Statistics at the University of Manitoba. He was a post-doctoral research fellow at Universities of Dortmund and Hannover, Germany, and an assistant professor at University of Basel, Switzerland, and a visiting scholar at University of Southern California, University of California - Berkeley, and University of Toronto.

报告题目Variable selection and estimation in quadratic measurements regression models with predictor measurement error

报告摘要:In many real data problems, e.g., in compressive sensing, signal processing and imaging, the response variables are nonlinear functions of the unknown parameters. In addition, many real data applications involve predictors that cannot be measured precisely. In the first part of this talk I will introduce a so-called quadratic measurements regression model that extends the usual linear model. We study the Lq regularized least squares method for variable selection and establish its weak oracle property. Moreover, we derive a fixed point equation and use it to construct an algorithm for numerical optimization. Numerical examples are given to demonstrate the performance of the proposed method and the efficiency of the algorithm. In the second part of the talk I will explore the impact of predictor measurement error on the selection and propose an instrumental variable method to correct the bias.