Hi everyone,
I encounter a problem regarding 2SLS in Stata and hope you could kindly give me some advice.
My data is panel data, and my regression model is y=β0+ β1*x1 + β2*x1^2 + β3*x2 +β4*control variables+.....
x1 and x2 are endogenous.
When using the stata code:
xi: ivreg dependent variable, exogenous variables,(endogenous variables=instruments)
I don't know how to deal with the quadratic term x1^2. I have IVs for both x1 and x2 based on previous literature, but not have an IV for the quadratic term. However, the "xi:ivreg" code requires each endogenous variable has a corresponding instrument. Is there any way to solve this problem, for example by using (predicted x1)^2? And what is the code I should use?
I have been struggling with this problem for a long time. Really hope someone can help me!
Thanks!
I encounter a problem regarding 2SLS in Stata and hope you could kindly give me some advice.
My data is panel data, and my regression model is y=β0+ β1*x1 + β2*x1^2 + β3*x2 +β4*control variables+.....
x1 and x2 are endogenous.
When using the stata code:
xi: ivreg dependent variable, exogenous variables,(endogenous variables=instruments)
I don't know how to deal with the quadratic term x1^2. I have IVs for both x1 and x2 based on previous literature, but not have an IV for the quadratic term. However, the "xi:ivreg" code requires each endogenous variable has a corresponding instrument. Is there any way to solve this problem, for example by using (predicted x1)^2? And what is the code I should use?
I have been struggling with this problem for a long time. Really hope someone can help me!
Thanks!
