部分线性变系数模型的一种新的轮廓(Profile)最小二乘估计  

A Novel Profile Least Square Estimation for the Partially Linear Varying-Coefficient Models

作　　者：张波[1] 郭海兵[1,2] 

机构地区：[1]中国人民大学统计学院,北京100872 [2]淮海工学院理学院,江苏连云港222005

出　　处：《数理统计与管理》2015年第2期275-283,共9页Journal of Applied Statistics and Management

基　　金：国家自然科学基金资助(71271210;71471173);教育部人文社会科学重点研究基地项目资助(14JJD910002)

摘　　要：作为部分线性模型和变系数模型的推广,部分线性变系数模型以其良好的适应性和稳健性受到了广泛的关注。本文基于函数的局部线性拟合,给出部分线性变系数模型的另一种轮廓(profile)最小二乘估计的方法,并从理论上证实了所得估计量具有良好的渐近性质,最后给出了估计方法的实例分析。The partially linear varying-coefficient model,as an extension of the partially linear model and the varying coefficients model,has been received increasing attention due to their good adaptability and robustness.In this paper,a novel profile least square estimation method for the partially linear varying coefficient models is proposed and asymptotic properties for the estimator s are obtained.Additionally,an application about CD4 data by using the proposed method is proceed.

关 键 词：部分线性变系数模型 变系数模型 局部多项式估计 渐近性 CD4数据 

分 类 号：O212.7[理学—概率论与数理统计]