How do you validate the assumptions of a causal inference model?
Instruction: Describe methods to assess whether the key assumptions of a causal model hold in empirical data.
Context: This question seeks to gauge the candidate’s ability to critically evaluate the validity of causal models by testing underlying assumptions.
Updated
Official answer available
Read the opening below, then unlock the full answer and practical guidance.
I would separate assumptions we can investigate empirically from those that observed data cannot prove. I can check treatment overlap, inspect extreme weights, verify timing and measurement, and assess balance in measured pretreatment covariates...
Related Questions
-
medium
-
hard
-
hard
-
hard
-
hard