题目：Variable Selection and Estimation in Regression Models with Quadratic Measurements and Predictor Measurement Error

报告人：Liqun Wang（University of Manitoba）

Abstract:

Regularization methods for high dimensional variable selection and estimation have been intensively studied in recent years and most of them are developed in the framework of linear regression models. In many real data problems, e.g. compressive sensing, Signal processing and imaging, the response variables are nonlinear function 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 quadratic measurements regression model that extends the usual linear model. We study the Lq regularized Ieast 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 selection bias in linear models.

时间：5月21日（周一）下午4:00

地点：首都师大新教2楼  813 教室

 欢迎全体师生积极参加！