Hi,
This is a follow-up on my previous e-mail which was perhaps not clear enough. My problem is not that I include a lagged value of the dependent variable as regressor in fixed estimation, but rather that I use a lag of an explanatory variable as an additional regressor. That is, I regress industrial productivity on proportion of the workforce trained this year, as well as on the proportion of the workforce the year before using the command xtreg, fe.
Stata does not allow me to do this directly; the error message is that time-series operators are not allowed. When I generate a new variable for proportion of the workforce trained a year before, and thus do not use the lag-specification, my estimation goes through.
My question is: am I doing something 'illegal' by recoding my variable and making the estimation go through; are there inherent econometric problems with using the lag of an explanatory variable as an additional regressor?
Yours,
Bob Rijkers
*
* 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/
