NA样本下随机设计情形部分线性模型的经验似然推断  

Empirical Likehood for Partially Linear Models with Random Designs Under Negatively Associated Samples

作　　者：黄玉[1] 晏振 HUANG Yu;YAN Zhen(College of Education,Nanning University,Nanning 530200,China;College of Mathematics and Statistics,Guangxi Normal University,Guilin 541004,China)

机构地区：[1]南宁学院教育学院,广西南宁530200 [2]广西师范大学数学与统计学院,广西桂林541004

出　　处：《应用数学》2023年第4期940-950,共11页Mathematica Applicata

基　　金：国家自然科学基金(11901124);南宁学院2021年度教授培育工程项目(2021JSGC07)。

摘　　要：本文研究NA样本下随机设计情形部分线性模型的经验似然推断,将分块技术应用到经验似然方法中,证明部分线性模型的参数β的对数经验似然比统计量的渐近分布为卡方分布,由此构造了NA样本下β的经验似然置信区间.同时,在有限样本情况下给出运用分块技术的经验似然与不运用分块技术的经验似然的模拟对比结果.In this paper,we study the blockwise empirical likelihood(EL)method to construct the confidence regions for the regression vectorβin a partially linear model with random designs under negatively associated samples.It is shown that the blockwise EL ratio statistic forβis asymptoticallyχ^(2) distributed,which are used to construct the EL-based confidence region forβwith random designs under negatively associated samples.Then,results of a simulation study on the finite sample performance of the empirical likelihood with block technique and the empirical likelihood without block technique are compared.

关 键 词：部分线性模型 随机设计 分块经验似然 NA样本 置信区间 

分 类 号：O212.7[理学—概率论与数理统计]