1 / 6 Causal Inference

What happens when causal overlap becomes weak?

In observational causal estimation, estimated propensity scores are extremely close to zero for some treated observations and extremely close to one for some controls. Which interpretation is most accurate?

A

Cross-fitting automatically restores positivity because each propensity model is trained on fewer observations

B

Identification remains unaffected because extreme propensity scores indicate highly predictable treatment assignment

C

Only outcome-regression estimators are affected; weighting estimators remain stable after normalisation

D

Inverse-probability weights can become unstable, and trimming may reduce variance while changing the effective target population