交互作用偏差对双系统估计量的影响

发布时间：2018-04-23 03:34

本文选题：性别比率 + 交互作用偏差　；参考：《数理统计与管理》2017年04期

【摘要】：虽然双系统估计量目前是人口普查质量评估领域估计总体实际人口数的主要方法,但其内在固有的缺陷却依然存在,即由于人口普查与其质量评估调查不独立引起的交互作用偏差使其低估或高估人口数。独立性假设失败源于在普查及其质量评估调查中登记的因果相关性,以及在普查及质量评估调查中登记概率的异质性。Bell模型是当前公认的测算交互作用偏差的有效方法。该方法建立在0-17岁及成年女性的双系统估计值不存在交互作用偏差的假设条件下。利用美国普查局提供的2010年资料全面展示了双系统估计量交互作用偏差的测算过程。实证结果表明,黑人成年男性的双系统估计值存在显著的交互作用偏差。研究有助于我国在未来人口普查质量评估工作中意识到交互作用偏差的存在,把测算的交互作用偏差添加到双系统估计量估计的人口数中,并依据修正后的双系统估计值计算人口普查净误差。
[Abstract]:Although dual-system estimates are currently the main method of estimating the overall actual population in the field of census quality assessment, their inherent shortcomings remain, That is, the population is underestimated or overestimated due to the interaction bias caused by the independence of the census and its quality assessment survey. The failure of independence hypothesis stems from the causality of registration in census and quality assessment survey, and the heterogeneity of registration probability in census and quality assessment survey. Bell model is widely accepted as an effective method to measure the deviation of interaction. The method is based on the assumption that there is no interaction bias in the two-system estimates of women aged 0-17 and adults. Based on the data provided by the U.S. Census Bureau in 2010, the calculation process of the interaction bias between the two system estimators is presented. The empirical results show that there is a significant interaction bias between the two-system estimates of black adult males. The research will help our country to realize the existence of the interaction deviation in the future census quality assessment, and add the calculated interaction deviation to the population estimated by the double system estimator. The net census error is calculated based on the modified double system estimate.
【作者单位】：重庆工商大学经济社会应用统计重庆市重点实验室;重庆工商大学长江上游经济研究中心;
【基金】：2016年重庆市社会科学规划重大委托项目(2016WT03) 国家社会科学基金项目(15BTJ011) 全国统计科学研究重点项目(2016LZ29) 重庆工商大学经济社会应用统计重庆市重点实验室资助项目重庆市社会科学规划年度项目(2016YBJJ022) 全国统计科学研究一般项目(2016LY28)
【分类号】：O212.1

【参考文献】