基于双系统估计量的中国非普查年人口总数估计  

Estimate Chinese Total Population in Noncensus Year Based on Dual System Estimator

作　　者：孟杰[1] 杨贵军[1] 

机构地区：[1]天津财经大学中国经济统计研究中心

出　　处：《数理统计与管理》2018年第2期298-308,共11页Journal of Applied Statistics and Management

基　　金：国家社会科学基金青年项目(17CTJ002);全国统计科学研究项目(2017LZ30);重庆市社会科学规划重大委托项目(2016WT03);国家自然科学基金面上项目(114711239)

摘　　要：针对当前中国非普查年人口总数估计方法的不足，本文提出利用人口抽样调查及其事后质量评估调查数据，构造基于捕获再捕获模型的非普查人口总数双系统估计量。首先，介绍双系统估计量的构造及其估计方法。其次，研究如何利用刀切法构建双系统估计量的方差估计。最后，通过一组模拟数据演示非普查年人口总数双系统估计量及其方差估计的构造过程。本文的研究能显著提高中国非普查年人口总数估计的准确性和可靠性，并对未来中国人口抽样调查的方案设计提供理论指导。Based on the deficiency of the Chinese total population estimation method in noncensus year, this paper proposes duM system estimator to estimate the total population in noncensus year for improving the accuracy of Chinese total population, used population sampling survey and its post quality evaluation survey. Firstly, this paper introduces the construction and estimation method of dual system estimator. Secondly, using jackknife method to build variance estimation of dual system estimator is introduced. Finally, this paper uses simulate data to demonstrate the practice process of dual system estimator and its variance estimator. This research contributes to improve the accuracy of Chinese total population in noncensus year, and provide theoretical direction for the future Chinese population sampling survey design.

关 键 词：人口抽样调查 非普查年人口总数估计 双系统估计量 刀切法 

分 类 号：O212[理学—概率论与数理统计]