高维半参数变系数部分线性测量误差模型的经验似然校正  

Calibration of the empirical likelihood for high-dimensional semiparametric varying-coefficient partially linear errors-in-variables model

作　　者：何军伟 陈夏[1] HE Junwei;CHEN Xia(School of Mathematics and Information Science,Shaanxi Normal University,Xi′an 710119,China)

机构地区：[1]陕西师范大学数学与信息科学学院,陕西西安710119

出　　处：《纺织高校基础科学学报》2018年第2期191-196,共6页Basic Sciences Journal of Textile Universities

基　　金：国家自然科学基金(11201276);陕西省自然科学基础研究计划项目(2014JQ1042);中央高校基本科研业务费专项基金(GK201503015)

摘　　要：针对含测量误差高维数据半参数变系数部分线性模型的经验似然校正问题,通过对对数经验似然比统计量统计性质的分析与计算,提出一种新的估计对数经验似然比统计量的期望和方差的方法,得到校正的对数经验似然比统计量的渐近分布,说明其渐近统计性质优于传统方法.并利用数值模拟证明了所提方法的优良性质.Aiming at the problem of the empirical likelihood calibration for high-dimensional semiparametric varying-coefficient partially linear errors-in-variables model,through analysis of the statistical properties of the empirical log-likelihood ratio and calculation,a new method to estimate the expectation and variance of the empirical log-likelihood ratio is given.The asymptotic distribution of the corrected empirical log-likelihood ratio is obtained,whose statistical properties are better than those of the traditional method.Simulation indicates that the proposed method performs well.

关 键 词：经验似然 渐近正态 校正 覆盖率 

分 类 号：O212.1[理学—概率论与数理统计]