Re: st: Quantile regression with fixed effects and first differences

[Gordon Hughes <G.A.Hughes@ed.ac.uk>]

It doesn't matter that you are trying to replicate another study if 
the specification doesn't make sense.  From your description, you 
have a panel dataset with a large number of individuals (panels) 
observed in a number of years.  You have taken first differences 
within each panel, which means that any individual fixed effects have 
been removed.

Why then, do you want to estimate a model with individual fixed 
effects?  This is equivalent to assuming that each individual unit 
displays some fixed growth component (linear or logarithmic depending 
on your dependent variable) over all years.  This is a pretty odd 
specification, especially as your year fixed effects imply that there 
is some systematic component of growth rates for different years.

If you really want to have individual fixed effects in growth rates, 
the obvious way round your problem is to take first differences of 
fDepVar, but I doubt whether that is a sensible model.

Gordon Hughes
g.a.hughes@ed.ac.uk 

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/