Firm effects in estimating earning function

[<Patrick.Heer@bakom.admin.ch>]

Dear all
This has nothing to do with R coding, but maybe someone can help me.
I am now dealing with dummy variables (8 firm indicators) in order to
regress the firm effects on the a wage distribution. My problem is that
there is always a reference category which I have to compare with the
coefficient estimated.
I have read a bench of workingpaper and it looks like there is a way to
compute all the firm coefficients without refering to one firm, since
every category shows up in the result with a coefficient value.
Can anyone help me with this issue?
Thank you
Patrick