英文论文
-
文献类型
-
Review
-
题名
-
Overlap weight and propensity score residual for heterogeneous effects: A review with extensions
-
作者
-
Choi, Jin-young; Lee, Myoung-jae
-
作者单位
-
[Choi, Jin-young] Xiamen Univ, Sch Econ, Xiamen, Peoples R China. [Lee, Myoung-jae] Korea Univ, Dept Econ, Seoul 02841, South Korea.
-
通讯作者地址
-
Korea Univ, Dept Econ, Seoul 02841, South Korea.
-
Email
-
jychoi@xmu.edu.cn; myoungjae@korea.ac.kr
-
ResearchID
-
-
ORCID
-
-
期刊名称
-
JOURNAL OF STATISTICAL PLANNING AND INFERENCE
-
出版社
-
ELSEVIER
-
ISSN
-
0378-3758
-
出版信息
-
2023-01, 222:22-37.
-
JCR
-
-
影响因子
-
-
ISBN
-
-
基金
-
National Research Foundation of Korea (NRF) - Korea government (MSIT) [2020R1A2C1A01007786]; Korea University fund [K2203251]
-
会议名称
-
-
会议地点
-
-
会议开始日期
-
-
会议结束日期
-
-
关键词
-
Overlap weight; Propensity score residual; Matching; Inverse probability weighting; Regression adjustment
-
摘要
-
Individual responses to a treatment D = 0, 1 differ, depending on covariates X. Averaging such a heterogeneous effect is usually done with the density of X, but averaging with 'overlap weight (OW)' is also often done, where OW is the normalized version of PSx(1-PS) with PS denoting the propensity score. OW attains its maximum at PS = 0.5, i.e., when subjects in one group have the best overlap with the other group, and OW accords several advantages to treatment effect estimators as reviewed in this paper. First, matching with OW addresses the non-overlapping support problem in a built-in way, without an arbitrary user intervention. Second, inverse probability weighting with OW overcomes the "too small denominator problem ", and can be efficient as well. Third, regression adjustment with OW is robust to misspecified outcome regression models. Fourth, covariate balance holds exactly, if OW is estimated by the generalized method of moment. In these advantages, the PS residual 'D - PS' plays a central role. We also discuss some shortcomings of OW, and show how seemingly unrelated estimators are in fact closely related through OW. Finally, we provide an empirical illustration.(C) 2022 Elsevier B.V. All rights reserved.
-
一级学科
-
Statistics & Probability
-
WOS入藏号
-
WOS:000814583400002
-
EI收录号
-
-
DOI
-
10.1016/j.jspi.2022.04.003
-
ESI
-
-
收录于
-
SCIE
-
返回