基于双系统估计量的人口普查净误差估计

doi:10.12341/jssms19398

系统科学与数学 ›› 2022, Vol. 42 ›› Issue (3): 676-696.DOI: 10.12341/jssms19398

基于双系统估计量的人口普查净误差估计

胡桂华¹, 叶宝红¹, 漆莉²

1. 重庆工商大学数学与统计学院, 经济社会应用统计重庆市重点实验室, 重庆 400067;
2. 重庆工商大学长江上游经济研究中心, 重庆 400067

收稿日期:2019-09-29 修回日期:2021-11-11 出版日期:2022-04-20 发布日期:2022-04-20
通讯作者: 叶宝红,Email:ybhysz123@163.com.
基金资助:
2021年国家社科基金后期资助暨优秀博士论文一般项目（21FTJB002）资助课题.

PDF ( 20 ) 引用

胡桂华, 叶宝红, 漆莉. 基于双系统估计量的人口普查净误差估计[J]. 系统科学与数学, 2022, 42(3): 676-696.

HU Guihua, YE Baohong, QI Li. Estimation of Net Census Coverage Error Based on Dual System Estimator[J]. Journal of Systems Science and Mathematical Sciences, 2022, 42(3): 676-696.

Estimation of Net Census Coverage Error Based on Dual System Estimator

HU Guihua¹, YE Baohong¹, QI Li²

1. School of Mathematics and Statistics, Chongqing Key Laboratory of Economic and Social Applied Statistics, Chongqing Technology and Business University, Chongqing 400067;
2. Economic Research Center of the Upper Reaches of the Yangtze-River, Chongqing Technology and Business University, Chongqing 400067

Received:2019-09-29 Revised:2021-11-11 Online:2022-04-20 Published:2022-04-20

针对中国和其他许多国家在使用双系统估计量时所存在的未对总体人口事后分层和未估计小区域净误差等诸多问题，从理论和实际操作层面构造基于捕获-再捕获模型的双系统估计量及合成双系统估计量.采用数理模型分析技术和抽样推断方法，研究双系统估计量，双系统估计量的抽样方差估计，以及小区域的净误差估计.研究结果表明：双系统估计量须在事后层建立，否则低估或高估总体实际人数；双系统估计量属于复杂估计量，分层刀切法适合于近似计算其抽样方差；不能忽视小区域的净误差估计.研究创新是，设计了适合于中国人口特点的事后分层方案，为科学有效地使用双系统估计量奠定了理论基础.研究价值在于，所构造的双系统估计量及其合成双系统估计量有望应用于中国2030年人口普查净误差估计.为便于读者理解，通过模拟，全面演示双系统估计量及合成双系统估计量的计算过程.

关键词： 政府统计, 抽样调查, 抽样方差

In response to many problems in the use of dual-system estimators in China and many other countries, such as failure to post-stratify the population and failure to estimate net error of small areas, a dual-system estimator based on the capture-recapture model and a synthetic dual-system estimator are constructed from the theoretical and practical levels. Dual-system estimator, its sampling variance estimation, and net error estimation of small areas, are studied by mathematical model analysis techniques and sampling inference methods. The research results show that:The dual-system estimator must be established in a post-stratum, otherwise the true number of people is underestimated or overestimated; the dual-system estimator is a complex estimator, and the stratified Jack-Knife method is suitable for approximate calculation of its sampling variance; the net error estimation of small areas cannot be ignored. The research innovation is that a post-stratification scheme which is suitable for the characteristics of China's population is designed to lay a theoretical foundation for scientific and effective use of the dual-system estimator. The research value is that the constructed dual-system estimator and the synthetic dual-system estimator are expected to be applied in the estimation of 2030 net error in China. To facilitate readers' understanding, calculation processes of the dual-system estimator and synthetic dual-system estimator are fully demonstrated through a simulation.

Key words: Governmental statistics, sample survey, sampling variance

MR(2010)主题分类:

62D05
62E20

分享此文:

[1] 胡桂华, 武洁, 丁杨. 人口普查质量评估中Logistic回归模型的应用. 数量经济技术经济研究, 2015, 32(4):106-122. (Hu G H, Wu J, Ding Y. Application of logistic regression model in census quality evaluation. The Journal of Quantitative & Technical Economics, 2015, 32(4):106-122.)
[2] 胡桂华, 余鲁, 丁杨. 人口普查净误差估计中的双系统估计量研究. 数量经济技术经济研究, 2016, 33(8):145-161. (Hu G H, Yu L, Ding Y. Research on the dual system estimator in the estimation of population census net error. The Journal of Quantitative & Technical Economics, 2016, 33(8):145-161.)
[3] Coale A. The population of the United States in 1950 classified by age, sex, and color a revision of census figures. Journal of the American Statistical Association, 1955, 50(269):16-54.
[4] Himes C, Clogg C. An overview of demographic analysis as a method for evaluating census coverage in the United States. Population Index, 1992, 58(4):587-607.
[5] 胡桂华, 吴东晟. 人口普查质量评估调查的抽样设计. 数量经济技术经济研究, 2014, 31(4):113-129. (Hu G H, Wu D S. Sample design of census quality evaluation survey. The Journal of Quantitative & Technical Economics, 2014, 31(4):113-129.)
[6] Hogan H, Wolter K. Measuring accuracy in a post-enumeration survey. Survey Methodology Journal, 1988, 14(1):99-116.
[7] 胡桂华, 杜艾卿. 基于单系统估计量的人口普查内容误差估计. 数理统计与管理, 2018, 37(6):951-963. (Hu G H, Du A Q. Estimation of census content error based on single system estimator. Journal of Applied Statistics and Management, 2018, 37(6):951-963.)
[8] 胡桂华, 丁宣浩, 陈义安, 等. 人口普查覆盖误差估计量的研究. 数理统计与管理, 2018, 37(1):1-12. (Hu G H, Ding X H, Chen Y A, et al. Research on census coverage error estimator. Journal of Applied Statistics and Management, 2018, 37(1):1-12.)
[9] 孟杰, 杨贵军. 基于双系统估计量的中国非普查年人口总数估计. 数理统计与管理, 2018, 37(2):298-308. (Meng J, Yang G J. Estimate Chinese total population in nocensus year based on dual system estimator. Journal of Applied Statistics and Management, 2018, 37(2):298-308.)
[10] 胡桂华, Robert M, Lara C. 人口普查净误差估计中的三系统估计量研究. 统计研究, 2017, 34(6):3-15. (Hu G H, Robert M, Lara C. Research on triple-system estimator in the estimation of net error in the census. Statistical Research, 2017, 34(6):3-15.)
[11] 胡桂华, 薛婷. 中国户籍登记系统覆盖评估研究. 统计与信息论坛, 2018, 33(7):34-46. (Hu G H, Xue T. Household registration system coverage evaluation research of China. Statistics and Information Forum, 2018, 33(7):34-46.)
[12] 史龙梅, 徐蔼婷. 西班牙"组合模式"人口普查:经验及启示. 统计与信息论坛, 2019, 34(4):32-40. (Shi L M, Xu A T. Interpretation of the Spain "population census combine use of sample survey and register-based survey" and its enlightenment to China. Statistics and Information Forum, 2019, 34(4):32-40.)
[13] 孟杰, 沈文静. 人口名录库及其在人口普查中的应用. 统计与信息论坛, 2018, 33(10):90-97. (Meng J, Shen W J. Population register-based and its application in census. Statistics and Information Forum, 2018, 33(10):90-97.)
[14] Griffin R. Potential uses of the administrative records for tripe-system modeling for estimation of census coverage error in $2020$. Journal of Official Statistics, 2014, 30(2):177-189.
[15] 胡桂华. 人口普查净误差估计综述. 数理统计与管理, 2018, 37(5):796-814. (Hu G H. Review for the census net error estimation. Journal of Applied Statistics and Management, 2018, 37(5):796-814.)
[16] Statistical Division of the United Nations. Post-Enumeration Surveys-Operational Guidelines. New York:United Nations Statistics Division, 2010.
[17] Goudie I B J, Goudie M. Who captures the marks for the Petersen estimator. Journal of the Royal Statistical Society:Series A (Statistics in Society), 2007, 170(3):825-839.
[18] Lincoln F. Calculating Waterfowl Abundance on the basis of banding returns. Circular of the Department of Agriculture, 1930, 118(6):1-4.
[19] Laplace P S. Sur les naissances, les mariages et les morts. Histoire de l'académie royale des sciences, 1783, 693-702.
[20] Sekar C C, Deming W E. On a method of estimating birth and death rates and extent of registration. Journal of the American Statistical Association, 1949, 44(245):101-115.
[21] Marks E, Krotki S W. Population growth estimation:A handbook of vital statistics measurement. The Population Council, 1974.
[22] 胡桂华. 美国2000年和2010年人口普查质量评估方法解读. 数理统计与管理, 2010, 29(2):262-276. (Hu G H. Interpretation of the quality assessment methods for the U.S. census 2000 and 2010. Journal of Applied Statistics and Management, 2010, 29(2):262-276.)
[23] 胡桂华. 人口普查质量评估中抽样后分层变量的选择. 数理统计与管理, 2015, 34(2):254-263. (Hu G H. Stratification variables selection after sampling for census quality assessment. Journal of Applied Statistics and Management, 2015, 34(2):254-263.)
[24] 孟杰. 双系统估计量的交互作用偏差研究. 数理统计与管理, 2019, 38(5):858-872. (Meng J. Research on correlation bias for dual system estimator. Journal of Applied Statistics and Management, 2019, 38(5):858-872.)
[25] 金勇进. 抽样:理论与应用. 北京:高等教育出版社, 2010. (Jin Y J. Sampling:Theory and Application. Beijing:Higher Education Press, 2010.)
[26] 冯士雍, 倪加勋, 邹国华. 抽样调査理论与方法. 北京:中国统计出版社, 2012. (Feng S Y, Ni J X, Zou G H. Sample Survey Theory and Method. Beijing:China Statistics Press, 2012.)
[27] McCaa R, 胡桂华, 廖金盆. 人口普查内容误差评估. 系统科学与数学, 2019, 39(11):1870-1884. (McCaa R, Hu G H, Liao J P. Evaluation of census content error. Journal of Systems Science and Mathematical Sciences, 2019, 39(11):1870-1884.)
[28] 胡桂华, 范署姗, 吴婷. 基于设计效应的人口普查质量评估调查样本量测算. 统计与信息论坛, 2020, 35(10):12-21. (Hu G H, Fan S S, Wu T. Sample size measurement of census quality survey based on design effect. Statistics and Information Forum, 2020, 35(10):12-21.)
[29] 刘礼, 邹国华. 缺失数据下Jackknife方差估计量的渐近设计无偏性. 系统科学与数学, 2006, 26(4):491-503. (Liu L, Zou G H. Asymptotically unbiased design of jackknife variance estimators with missing data. Journal of Systems Science and Mathematical Sciences, 2006, 26(4):491-503.)
[30] 邹国华, 冯士雍. 广义比估计与广义差估计及其优良性. 系统科学与数学, 1998, 18(3):359-365. (Zou G H, Feng S Y. Generalized ratio estimation, generalized difference estimation and their advantages. Journal of Systems Science and Mathematical Sciences, 1998, 18(3):359-365.)
[31] 李莉莉, 冯士雍, 秦怀振. 不放回样本追加策略下域的估计. 统计研究, 2007, 24(6):80-85. (Li L L, Feng S Y, Qin H Z. Estimation of the domain under the strategy of appending samples without replacement. Statistical Research, 2007, 24(6):80-85.)
[32] 胡桂华, 漆莉, 吴婷,等. 基于比率估计量的人口普查内容误差估计.工程数学学报, 2018, 35(6):622-634. (Hu G H, Qi L, Wu T, et al. Error estimation of census content based on ratio estimators. Chinese Journal of Engineering Mathematics, 2018, 35(6):622-634.)

[1]	金勇进, 刘晓宇. 大数据背景下的抽样调查[J]. 系统科学与数学, 2022, 42(1): 2-16.
[2]	杨昊宇, 秦祎辰, 李扬. 问卷分割设计的成组序贯子问卷分配法[J]. 系统科学与数学, 2022, 42(1): 17-34.
[3]	石峻驿. 大数据背景下平台类企业开展抽样调查的应用研究[J]. 系统科学与数学, 2022, 42(1): 100-108.
[4]	张璇, 赵静, 丁文兴. 大数据背景下产品质量抽样调查的样本量设计[J]. 系统科学与数学, 2022, 42(1): 133-140.
[5]	Robert McCaa，胡桂华，廖金盆. 人口普查内容误差评估[J]. 系统科学与数学, 2019, 39(11): 1870-1884.
[6]	邹国华;冯士雍. 超总体模型下有限总体的估计[J]. 系统科学与数学, 2007, 27(1): 27-38.
[7]	冯士雍. 系统抽样时方差估计量的比较──一种新的比较准则[J]. 系统科学与数学, 1998, 18(1): 47-057.

阅读次数

全文

摘要