Comparer des méthodes

	Analyse de sensibilité spatiale pour la causalité	Modèle d'erreur spatiale (SEM)
Domaine	Inférence causale	Analyse spatiale
Famille	Regression model	Regression model
Année d'origine	1988–2021 (developed progressively)	1988
Auteur d'origine	Anselin (1988) for spatial diagnostics; Reich et al. (2021) for spatial causal frameworks	Anselin
Type	Sensitivity / robustness analysis	Spatial regression (spatially autocorrelated errors)
Source fondatrice	Anselin, L. (1988). Spatial Econometrics: Methods and Models. Kluwer Academic Publishers, Dordrecht. ISBN: 978-9024737322	Anselin, L. (1988). Spatial Econometrics: Methods and Models. Kluwer Academic.
Alias	spatial causal sensitivity, spatial robustness checks, SSAC, spatial confounding sensitivity	SEM, spatial error regression, spatial autoregressive error model, Uzamsal Hata Modeli (SEM / Spatial Error)
Apparentées	6	5
Résumé	Spatial sensitivity analysis for causality systematically tests whether a causal estimate derived from georeferenced data holds up as spatial structure, spillovers, and the choice of spatial weights matrix are varied. Because nearby units often share unmeasured confounders — soil quality, local infrastructure, neighbourhood norms — a naive regression may yield biased causal estimates. This method reveals how fragile or robust a claimed causal effect is to alternative spatial specifications.	The Spatial Error Model, developed within Anselin's spatial econometrics framework (1988), is a regression model that assumes spatial dependence enters through the error term: the disturbances of neighbouring units are correlated. It is used when unobserved shared factors make the errors of nearby observations move together, and it is estimated by maximum likelihood or GMM rather than ordinary least squares.