Hello, I see on the systemfit manual that you can estimate one-equation IV - I have a variable, and need to test if it's endogeneous, but do not need to estimate a system. Does anyone have any examples of this? Do you just run OLS with the endogenous variable, and then run a Hausmann to test endogeneity of OLS resid. vs. IV resid? Thanks in advance, DM [[alternative HTML version deleted]]
Possibly Parallel Threads
- Systemfit with structural equations and cross equation parameter interaction
- IV estimation
- Problems with obtaining t-tests of regression coefficients applying consistent standard errors after run 2SLS estimation. Clearer !!!!!
- How to still processing despite bug errors?
- Hausman Test