摘要
依据一定的抽样方法在动态总体中选定的样本,在前后两期调查中会出现部分样本丢失的情况,从而导致部分样本调查数据的缺失,如果直接忽视丢失样本信息,则会降低样本的有效性,影响总体参数估计的精度。文章针对动态总体分层抽样中前后两期样本出现丢失的情况,将前期样本信息在第二期内消亡和新生的样本视为不可观测样本,采用多重插补技术对丢失样本目标变量值进行估计,设计出基于多重插补技术的第二期分层样本的参数估计方法,并构造相应的估计量。实验表明,基于多重插补技术的分层抽样具有优良性,设计的参数估计量具有无偏性,且较为有效。
According to a certain sampling method, some samples selected in the dynamic population will be lost in the two surveys, which leads to the lack of survey data of some samples. If the lost sample information is ignored directly, the validity of the sample will be reduced and the accuracy of the overall parameter estimation will be affected.In view of the loss of samples in the two periods of stratified sampling of dynamic population, this paper regards the samples whose information of the previous period disappears in the second period as unobservable samples, and then uses the multiple interpolation technology to estimate the target variable value of the lost samples.The multi-interpolation technique is used to estimate the value of the target variable of the lost samples. The parameter estimation method of the layered samples in the second phase based on the multi-interpolation technique is designed, and the corresponding estimators are constructed. The experimental results show that the layered sampling based on multiple interpolation technique is superior, and the designed parameter estimation is unbiased and effective.
作者
张维群
段格格
Zhang Weiqun;Duan Gege(School of Statistics,Xi’an University of Finance and Economics,Xi’an 710100,China)
出处
《统计与决策》
北大核心
2023年第2期15-19,共5页
Statistics & Decision
基金
国家社会科学基金重点项目(20ATJ002)
国家社会科学基金重大项目(21&ZD147)。
关键词
动态总体
不可观测样本
多重插补
分层抽样估计
dynamic population
unobservable sample
multiple interpolation
stratified sampling estimation
