方法对比

	多期反事实概率加权法	动态逆概率加权
领域	因果推断	因果推断
方法族	Regression model	Regression model
起源年份	2000	1986-2000
提出者	Robins, Hernan & Brumback	James M. Robins and colleagues
类型	Weighted causal estimator	Causal weighting estimator
开创性文献	Robins, J. M., Hernan, M. A., & Brumback, B. (2000). Marginal structural models and causal inference in epidemiology. Epidemiology, 11(5), 550-560.	Robins, J. M., Hernan, M. A., & Brumback, B. (2000). Marginal structural models and causal inference in epidemiology. Epidemiology, 11(5), 550-560.
别名	longitudinal IPW, multi-period IPW, time-varying IPW, sequential IPW	Dynamic IPW, Time-varying IPW, Longitudinal IPW, Sequential IPW
相关	6	4
摘要	Multi-period Inverse Probability Weighting (IPW) estimates the causal effect of a treatment that varies across multiple time periods by reweighting observations according to the probability of receiving each period's treatment given past treatment history and time-varying confounders. It creates a pseudo-population where treatment at each period is independent of measured confounders, enabling unbiased estimation of sustained treatment strategies.	Dynamic Inverse Probability Weighting (Dynamic IPW) estimates the causal effect of a time-varying treatment sequence by reweighting observed data to mimic a hypothetical randomised trial. Developed by Robins and colleagues in the context of marginal structural models, it handles the challenge that in longitudinal settings, past treatment affects future covariates, which in turn affect future treatment — a feedback loop that standard regression cannot untangle.