黄振生:Estimation for varying coefficient partially nonlinear models with distorted measurement errors

时间:2017-10-27浏览:166设置

题目:Estimation for varying coefficient partially nonlinear models with distorted

measurement errors

报告人:黄振生教授南京理工大学

时间:113日(周五)上午10:00-11:00

地点:统计楼105  

摘要:In this paper, we propose a new varying coefficient partially nonlinear model when

both the response and predictors are not directly observed, but are observed by distorting

unknown functions of commonly observable covariate. Because of the complexity of models,

the existing estimation methods can not be directly employed. For this, we propose to

employ an efficient nonparametric regression to estimate the unknown distortion functions

concerning covariates and response variable, and further, we apply a profile nonlinear least

squares estimation procedure to estimate the parameters and the coefficient functions. We

also establish the asymptotic properties of all the proposed estimators. To illustrate our

proposed methodology, we carry out some stimulated and real examples. 

报告人简介:黄振生,南京理工大学统计与金融数学系教授、博导、系主任。2010年于华东师范大学获得博士学位;主要从事非参数统计及其应用等相关领域研究;主持国家自然科学基金2项,参与国家级和省部级项目多项;担任国际期刊AE及通讯评审人,担任国家自然科学基金委及教育部等部委的通讯评审专家,获得上海市优秀博士论文奖,澳大利亚联邦政府“奋进”奖,江苏省“青蓝工程”中青年学术带头人,迄今已发表SCI学术论文50余篇。

返回原图
/