Thanks, Marteen
that's true of course. However, and tell me if I do wrong, when I run a first stage regression of a variable that takes different values any time on a variable that has the same values over time, I get fitted values that are different over time (and I saw it in fact looking at the dataset). Hence, if I then regress these fitted values on the other variables using fixed effects, I do not understand why he drops them, while the fitted values are not constant over time, as other variables.
Thanks again for your and any help
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/
