0
$\begingroup$

I am trying to examine the determinants of economic growth, essentially regressing a number of variables upon economic growth. I face a problem that there is likely simultaneous causality between many of my explanatory variables and the dependent variable. To deal with it I am using instrumental variables. I was wondering would it be reasonable to use the past values of my explanatory variables as instruments for their current values?

Thank you so much for the answers.

$\endgroup$
0
$\begingroup$

I suggest looking into System GMM. It is a panel regression method that uses past values as instruments and is exactly what you're looking for, I think. It has a few restrictions that much be met, the most important requires you to have more cross-sections than time periods. Without seeing your model and other details that is all I can provide for help. I hope this works for you.

$\endgroup$

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.