Comparer des méthodes

	Modèle Structurel Marginal Robuste	Modèle structurel marginal (MSM) pour données de panel
Domaine	Inférence causale	Inférence causale
Famille	Regression model	Regression model
Année d'origine	2000–2004	2000
Auteur d'origine	Robins, Hernán & Brumback; robustness extensions by Scharfstein, Rotnitzky, Lunceford & Davidian	James M. Robins, Miguel A. Hernan, Babette Brumback
Type	Causal inference / weighted regression	Causal model for time-varying treatments
Source fondatrice	Robins, J. M., Hernán, 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.
Alias	robust MSM, doubly-robust MSM, sandwich-SE MSM, robust IPTW marginal structural model	MSM panel, longitudinal MSM, panel MSM, time-varying treatment MSM
Apparentées	6	5
Résumé	Robust Marginal Structural Models (robust MSMs) extend the standard MSM framework — which uses inverse probability of treatment weighting to handle time-varying confounding — by pairing IPTW estimation with sandwich (robust) standard errors or doubly-robust estimators. This combination yields valid causal estimates and reliable inference even when the outcome regression model is mildly misspecified or weights are moderately variable.	A panel data marginal structural model (MSM) uses inverse probability of treatment weighting (IPTW) across multiple time periods to estimate the causal effect of a time-varying treatment, while appropriately adjusting for time-varying confounders that are themselves affected by prior treatment — a bias source that conventional regression cannot handle.