Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Statistics & Decision》 2018-17
Add to Favorite Get Latest Update

Methods for Parameter Estimation With Missing Covariate Data

Yu Lichao;College of Science, Minzu University of China;  
In sample survey, how to estimate total parameters of data sets with missing data is a hot topic. At present, existing methods mainly focus on the situation of the data loss of dependent variables, but there are few studies on the situation of the lack of covariates. In the case that the data loss mechanism of covariate is MAR or NMAR, this paper introduces several methods of parameter estimation in the case of covariate loss, including multiple imputation method, Bayes method and maximum likelihood method, trying to use EM algorithm, Gibbs sampling algorithm, data expansion algorithm and other statistical calculation methods to solve the parameter estimation problem in the case of missing covariate. Finally through numerical simulation, the paper makes a comparison among several methods.
【Fund】: 国家社会科学基金青年项目(18CTJ011);; 全国统计科学研究项目重点项目(2017LZ01)
【CateGory Index】: O212
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved