Propensity Score
定义 Definition
倾向得分:在因果推断与观察性研究中,指“在给定一组协变量条件下,个体接受某种处理/干预(treatment)的概率”。常用于匹配、分层、加权等方法,以减少混杂偏倚、让处理组与对照组更可比。
发音 Pronunciation
/prəˈpɛn.sə.ti skɔːr/
例句 Examples
The researchers used propensity scores to match patients in the two groups.
研究人员使用倾向得分来匹配两组患者。
After estimating the propensity score with a logistic regression model, the team applied inverse-probability weighting to reduce confounding and estimate the treatment effect.
在用逻辑回归模型估计倾向得分之后,团队采用逆概率加权来减少混杂因素,并估计处理效应。
词源 Etymology
propensity 来自拉丁语 propensus(“倾向的、趋向于的”),表示“倾向/偏好”;score 原指“刻痕、记号”,后引申为“分数/评分”。“propensity score(倾向得分)”作为统计学与流行病学术语,常与 Rosenbaum 与 Rubin 的工作相关,用来把“接受处理的倾向”用一个概率数值概括,便于在观察性数据中进行调整与比较。
相关词 Related Words
文献与作品 Literary Works
- Rosenbaum, P. R., & Rubin, D. B. (1983). The central role of the propensity score in observational studies for causal effects.
- Stuart, E. A. (2010). Matching methods for causal inference: A review and a look forward.
- Austin, P. C. (2011). An introduction to propensity score methods for reducing the effects of confounding in observational studies.
- Hernán, M. A., & Robins, J. M. Causal Inference: What If(在线教材/专著,系统讨论倾向得分与加权等方法)
- Imbens, G. W., & Rubin, D. B. Causal Inference for Statistics, Social, and Biomedical Sciences(专著,涵盖倾向得分与因果推断框架)