手法を比較

	局所的平均処置効果（LATE / CACE）	フロントドア調整（フロントドア基準）
分野	因果推論	因果推論
系統	Regression model	Regression model
提唱年	1994	1995
提唱者	Imbens & Angrist (1994); Angrist, Imbens & Rubin (1996)	Judea Pearl
種類	Instrumental-variable causal estimand	Causal identification (graphical adjustment)
原典	Imbens, G. W., & Angrist, J. D. (1994). Identification and Estimation of Local Average Treatment Effects. Econometrica, 62(2), 467-475.	Pearl, J. (1995). Causal Diagrams for Empirical Research. Biometrika, 82(4), 669-688.
別名	LATE, CACE, complier average causal effect, Yerel Ortalama Tedavi Etkisi (LATE / CACE)	frontdoor criterion, Pearl's frontdoor adjustment, frontdoor formula, Ön Kapı Düzenlemesi (Frontdoor Adjustment)
関連	5	4
概要	The Local Average Treatment Effect is an instrumental-variable estimand, introduced by Imbens and Angrist (1994) and formalised with Rubin (1996), that recovers the average treatment effect for the subpopulation of compliers — units whose treatment status is actually moved by the instrument. It is closely tied to compliance analysis.	Frontdoor adjustment is Judea Pearl's graphical identification strategy, introduced in 1995, that recovers the causal effect of a treatment on an outcome through a fully mediating variable even when an unobserved confounder sits between the treatment and the outcome. It is the go-to tool when the backdoor criterion cannot be satisfied because the confounder is unmeasured.