Hello everyone,
I am estimating IV for my binary endogenous independent variable by using ivpoisson. My command is like this: ivpoisson gmm Y X1 X2 X3 i.year (X4 = Z)
X4 is a binary endogenous variable. Instrumental variable Z is discrete.
My question is:
1. Am I using the right command to estimate IV for binary endogenous variables? Stata manual states that ivpoisson is only applicable for continuous endogenous variables. However, prof. Wooldridge once mentioned on this forum that it can be used for any kind of endogenous variable.
2. Are there other ways to conduct the 2SLS in this case? For example, probit in the first stage and ppml in the second?
3. What kinds of tests should I conduct after ivpoisson regression?
I read many posts but have not found a clear solution to this problem. I would really appreciate if anyone can help.
Many thanks.
I am estimating IV for my binary endogenous independent variable by using ivpoisson. My command is like this: ivpoisson gmm Y X1 X2 X3 i.year (X4 = Z)
X4 is a binary endogenous variable. Instrumental variable Z is discrete.
My question is:
1. Am I using the right command to estimate IV for binary endogenous variables? Stata manual states that ivpoisson is only applicable for continuous endogenous variables. However, prof. Wooldridge once mentioned on this forum that it can be used for any kind of endogenous variable.
2. Are there other ways to conduct the 2SLS in this case? For example, probit in the first stage and ppml in the second?
3. What kinds of tests should I conduct after ivpoisson regression?
I read many posts but have not found a clear solution to this problem. I would really appreciate if anyone can help.
Many thanks.
