Hi everyone,
I used two different sets of instrumental variables for regression (ivreg2), and the estimatiions are radically different. The coefficient of the key variable is positive for one set of IVs, but negative for the other set. both sets of IVs are valid according to the tests in 2sls. In such case, what should I do?
Thank you so much in advance for help!
I used two different sets of instrumental variables for regression (ivreg2), and the estimatiions are radically different. The coefficient of the key variable is positive for one set of IVs, but negative for the other set. both sets of IVs are valid according to the tests in 2sls. In such case, what should I do?
Thank you so much in advance for help!
