Dear all,
I have a panel of regional data for ten years. I want to instrument a variable and apply xtivreg, fe. However I have an instrument very old that is only available for one year. Hence, I thought it was possible to perform such an estimation, since the first stage regression gives different fitted values (of course,since the variable to be instrumented changes with time). However when I run such a regression, stata says that the regression is not identified. I tried also to perform myself the first step, that, as I supposed, gives different values for each year. But then, when I perform the fixed effects estimation with such fitted values, it drops them.
May anyone help me to understand why?
Many thanks
jenny
*
* 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/
