手法を比較

	機械学習拡張プラセボテスト	因果推論のための操作変数（IV）法
分野	因果推論	医療経済学
系統	Regression model	Process / pipeline
提唱年	2010s–2018	1990s (modern applications)
提唱者	Chernozhukov, Hansen, and collaborators; Athey and Imbens	Angrist & Pischke (applied econometrics); rooted in econometric theory
種類	Causal validation / falsification test	Method
原典	Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C., Newey, W., & Robins, J. (2018). Double/debiased machine learning for treatment and structural parameters. The Econometrics Journal, 21(1), C1-C68.	Angrist, J. D., & Pischke, J. S. (2009). Mostly Harmless Econometrics: An Empiricist's Companion. Princeton: Princeton University Press.
別名	ML placebo test, data-driven placebo falsification, ML-augmented falsification test, ML permutation placebo	IV, two-stage least squares, TSLS, causal estimation
関連	3	3
概要	The machine learning-augmented placebo test is a causal-inference validation technique that uses flexible ML estimators — such as causal forests, LASSO, or double/debiased ML — to conduct falsification checks on an identification strategy. By replacing real treatment assignments with placebo (fake) assignments and verifying that the estimated effect collapses to zero, researchers confirm that their causal findings are not artefacts of model misspecification or confounding.	Instrumental variables (IV) is an econometric method to estimate causal effects when treatment or exposure is not randomly assigned and confounding is severe or unmeasured. IV relies on a third variable (instrument) that influences treatment but does not directly affect the outcome, allowing researchers to isolate the causal effect from the noise of confounding. Developed extensively in econometrics (Angrist & Pischke, 1990s–2000s), IV methods are increasingly used in health economics and health services research to leverage natural experiments and policy changes.