THE CENTRAL ROLE OF THE PROPENSITY SCORE IN OBSERVATIONAL STUDIES FOR CAUSAL EFFECTS
THE CENTRAL ROLE OF THE PROPENSITY SCORE IN OBSERVATIONAL STUDIES FOR CAUSAL EFFECTS
复制标题
DOI：
10.1093/biomet/70.1.41
复制
发表时间：
1983-01-01
期刊：
影响因子：
2.7
通讯作者：
RUBIN, DB
中科院分区：
文献类型：
--
作者：
ROSENBAUM, PR;RUBIN, DB
倾向得分是给定观察到的协变量向量分配到特定处理的条件概率。大样本和小样本理论都表明，调整标量倾向得分足以消除所有观察到的协变量造成的偏差。应用包括：(i)在单变量倾向得分上进行匹配抽样，这是判别匹配的概括；（ii）在倾向得分上通过子分类进行多变量调整，其中使用相同的子分类来估计所有结果变量和所有亚群体的治疗效果；（iii）通过二维图直观地表示多变量协方差调整。
The propensity score is the conditional probability of assignment to a particular treatment given a vector of observed covariates. Both large and small sample theory show that adjustment for the scalar propensity score is sufficient to remove bias due to all observed covariates. Applications include: (i) matched sampling on the univariate propensity score, which is a generalization of discriminant matching, (ii) multivariate adjustment by subclassification on the propensity score where the same subclasses are used to estimate treatment effects for all outcome variables and in all subpopulations, and (iii) visual representation of multivariate covariance adjustment by a two- dimensional plot.