Covariate-balancing-propensity-score-based inference for linear models with missing responses

详细信息    查看全文

• 作者：Donglin Guo^a ; ^b ; ^{gdl1105@emails.bjut.edu.cn} ; Liugen Xue^a ; Yuqin Hu^a ; ^c
• 关键词：Linear model ; Missing at random ; Covariate balancing propensity score ; GMM ; Augmented inverse probability weighted ; Robust estimation
• 刊名：Statistics & Probability Letters
• 出版年：2017
• 出版时间：April 2017
• 年：2017
• 卷：123
• 期：Complete
• 页码：139-145
• 全文大小：409 K
• 卷排序：123

文摘

Based on covariate balancing propensity score (CBPS), improved estimators for the regression coefficients and population mean of linear models are obtained, when the responses are missing at random. It is proved that the proposed estimators are asymptotically normal.