Hi all
I am estimating a labour supply equation with wage as a LHS variable. When I use -xtivreg, fe- the logwage variable becomes insignificant with a very high standard error. However, if I perform the IV manually by first estimating a fixed effects wage equation and then using the log of the predicted wage in the fixed effects hours equation, the estimate is fine.
Note that the instrument used passes the tests of validity.
My questions are
1. Can anyone tell me why I am getting these different results?
2. How much should I be concerned about this given that the wage is not really the variable I am interested in?
Many thanks in advance for any help.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
