Condition for Instrumental Variable (IV) approach: random assignment

[Ralph Klijn]

Hello everyone,

I am pursuing an IV strategy in Stata 15. However, in my search for an IV, I stumbled on the requirement that the IV must be randomly assigned.

I do not, however, seem to fully understand why.

Is someone able to explain this to me?

Thank you in advance.

Best,
Ralph

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

I don't understand your question. I guess there might be experimental situations or something where random assignment might have a role, but I don't remember hearing about this in terms of 2SLS or other iv techniques. An iv must be correlated with the endogenous variable and not correlated with the errors of the primary equation.