เปรียบเทียบวิธี

	การจับคู่แบบแม่นยำที่ปรับหยาบเชิงพื้นที่ (Spatial CEM)	การประมาณค่าแบบทนทานสองชั้นเชิงพื้นที่
สาขาวิชา	การอนุมานเชิงสาเหตุ	การอนุมานเชิงสาเหตุ
ตระกูล	Regression model	Regression model
ปีกำเนิด	2012 (CEM foundation); spatial extension in applied literature 2015-present	2010s–2020s
ผู้ริเริ่ม	Iacus, King & Porro (CEM foundation, 2012); extended to spatial contexts by applied spatial econometricians	Extension of Robins, Rotnitzky & Zhao (1994) doubly robust framework to spatial settings; developed in spatial epidemiology and econometrics literature
ประเภท	Quasi-experimental matching estimator with spatial covariates	Semiparametric causal estimator
แหล่งต้นตำรับ	Iacus, S. M., King, G., & Porro, G. (2012). Causal Inference without Balance Checking: Coarsened Exact Matching. Political Analysis, 20(1), 1-24.	Papadogeorgou, G., Mealli, F., & Zigler, C. M. (2019). Causal inference with interfering units for cluster and population level treatment allocation programs. Biometrics, 75(3), 778-787.
ชื่อเรียกอื่น	Spatial CEM, Geographic CEM, Spatial exact matching, CEM with spatial covariates	Spatial DR, Spatial AIPW, Spatial augmented IPW, Doubly robust spatial causal estimation
ที่เกี่ยวข้อง	6	5
สรุป	Spatial Coarsened Exact Matching applies the Coarsened Exact Matching framework to study designs involving geographic units — neighbourhoods, census tracts, municipalities, or grid cells. Covariates are coarsened into discrete bins and units are matched exactly on those bins, with spatial attributes (location, adjacency, geographic characteristics) incorporated as matching dimensions to control for spatial confounding.	Spatial doubly robust estimation is a semiparametric causal inference method that combines propensity score weighting with outcome regression modeling — providing protection against misspecification of either component — while explicitly accounting for spatial autocorrelation among units. It extends the classical augmented inverse probability weighting (AIPW) estimator to settings where treatment assignment and outcomes are geographically clustered or spatially dependent.