摘要
候选者数据库网络调查的推断问题是网络调查发展中迫切需要解决的问题.基于此,提出基于贝叶斯伪设计与组合样本的非概率抽样推断方法:将网络候选者数据库的调查样本与概率样本结合,根据贝叶斯定理推导出网络候选者数据库的调查样本单元的伪权数构造式,再利用两个样本数据共同估计总体均值,最后利用Bootstrap和Jackknife方法来计算总体均值估计的方差估计.研究结果表明:基于贝叶斯伪设计与组合样本的总体均值估计比使用Elliot方法估计的总体均值偏差更小,估计效果较好;方差估计方面,Bootstrap方差估计比Jackknife方差估计的效果好.
How to solve the inference problem of candidate database web surveys is an urgent problem to be solved in the development of web survey.In order to solve this problem,the inference method of non-probability sampling based on Bayesian pseudo design and the combined sample is proposed.Firstly,a survey sample of the web candidate database and a probability sample are combined.The pseudo weight formula of the survey sample of the web candidate database is then derived according to the Bayesian theory.Furthermore,the data of the combined sample is used to estimate the population mean.Lastly,the bootstrap and jackknife methods are used to compute the variance estimator of the population mean estimator.The research results show that the population mean estimator based on Bayesian pseudo design and the combined sample is better,and has lower bias than the estimator using the Elliot’s approach.The variance estimator based on the bootstrap method is better than the variance estimator based on the jackknife method.
作者
刘展
金勇进
LIU Zhan;JIN Yongjin(Hubei Key Laboratory of Applied Mathematics, School of Mathematics and Statistics, HubeiUniversity, Wuhan 430062;Center for Applied, Statistics, Renmin University of China,Beijing 100872)
出处
《系统科学与数学》
CSCD
北大核心
2019年第6期990-1000,共11页
Journal of Systems Science and Mathematical Sciences
基金
国家社会科学基金项目(18BTJ022)资助课题
关键词
贝叶斯
伪设计
组合样本
网络候选者数据库
非概率抽样
Bayesian
pseudo design
combined sample
web candidate database
non-probability sampling
