吉林大学学报(理学版)

• 数学 • 上一篇    下一篇

多总体测量误差模型中的经验似然推断

袁晓惠1, 刘天庆2   

  1. 1. 长春工业大学 基础科学学院, 长春 130012; 2. 吉林大学 数学学院, 长春 130012
  • 收稿日期:2015-07-06 出版日期:2016-03-26 发布日期:2016-03-23
  • 通讯作者: 刘天庆 E-mail:tqliu@jlu.edu.cn

Empirical Likelihood Inference for MeasurementError Model with Multiple Populations

YUAN Xiaohui1, LIU Tianqing2   

  1. 1. School of Basic Science, Changchun University of Technology, Changchun 130012, China;2. Collge of Mathematics, Jilin University, Changchun 130012, China
  • Received:2015-07-06 Online:2016-03-26 Published:2016-03-23
  • Contact: LIU Tianqing E-mail:tqliu@jlu.edu.cn

摘要:

在多总体测量误差模型中, 运用经验似然方法得到总体参数和累积分布函数(CDF)的有效估计, 并将多总体经验似然方法应用于NS抽样模型, 得到了共同均值和CDF的有效估计. 结果表明, CDF的估计不仅渐近无偏, 而且利用了所有样本信息.

关键词: 累积分布函数, 有效, 经验似然, 测量误差, 多总体

Abstract:

We used the empirical likelihood method to obtain estimators of the population parameters and cumulative distribution function (CDF) for measurement error model with multiple populations. Our estimator of CDF is asymptotically unbiased and utilizes all the available sample information. Then we
applied the empirical likelihood to the NS sampling model and obtained the efficient estimation of the common mean and CDF. Simulation study further demonstrates the good properties of our estimators.

Key words: cumulative distribution function, efficient, empirical likelihood, measurement error, multiple populations

中图分类号: 

  • O212.4