I am working on a new causal inference method for estimating treatment effects and want to test it against current methods. I would like to find a dataset where the counterfactual is either observed in an almost identical unit (i.e. the same person at a later point in time, not a different person; I realize this is impossible for the exact same unit) or the set up allows for good estimation of the counterfactual. I've spent a while trying to brainstorm ideas, and would like some outside input. I think biological studies that involve genetically identical mice given and the Card-Krueger minimum wage data near the PA/NJ state line are potential examples.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.