On Oct 18, 2005, at 4:38 AM, Dirk Nachbar wrote:
Dear all
I want to do a panel regression where I have also a lag on the rhs
of the equation, but trying
xtreg y l.y
doesn't work. also i cannot generate a lag variable. how should i
do it?
thanks
Dirk Nachbar
You probably know this, but if you're using fixed effects, then
including a lag of the dependent variable is going to violate the
strict exogeneity assumptions required for consistent estimates. You
may want to look into Arrelano and Bond's dynamic GMM estimator, (-
xtabond- in Stata).
But, if you want to run -xtreg- with lagged dependent variables, then
you can always generate a variable for those lags. I think the
command would be - gen ly=[_n-1] - or something like that.
scott
*
* 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/