0
$\begingroup$

If using an identificaton strategy such as differences in differences, Regression Discontinuities, or IV's, I see the phrase 'as good as random' used alot. Does this have a rigorous definition, or is it just another way of saying what we are trying to accomplish when ensuring exogeneity of the above estimators?

$\endgroup$
  • $\begingroup$ "Does this have a rigorous definition...?" No. $\endgroup$ – Michael Aug 17 '20 at 0:12
0
$\begingroup$

The phrase "as good as random" is often just used instead of (conditional) igonorability (no open back-door paths) in observational natural experimental settings. In my opinion, this is bad practice and shows the authors insecurity of explicitly stating that the back-door criterion is assumed to hold - likely because they think it does not.

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.