I am running -eteffects- to estimate the (endogenous treatment) effect of X (endegenous) on Y, and using the exclusion restriction (IVs) as Z1 and Z2.
The reviewer has asked for Conley's plausibly exogenous approach to support the exclusion criteria. I plan to use -plausexog- and -imperfectiv- (both from SSC) for this. My questions:
1. In my setup, X is binary, Z1 is binary and Z2 is numeric. Do the above methods work when treatment and the IV is binary?
2. How to interpret the results after -plausexog- and -imperfectiv-?
3. Do these methods completely replace IV/Endogenous treatment effects methods or act as robustness check only?
The reviewer has asked for Conley's plausibly exogenous approach to support the exclusion criteria. I plan to use -plausexog- and -imperfectiv- (both from SSC) for this. My questions:
1. In my setup, X is binary, Z1 is binary and Z2 is numeric. Do the above methods work when treatment and the IV is binary?
2. How to interpret the results after -plausexog- and -imperfectiv-?
3. Do these methods completely replace IV/Endogenous treatment effects methods or act as robustness check only?
