Regress only on fixed effects (residualize) without using dummies

[David Tsu]

I want to regress on only fixed effects to obtain dipersion of within-variation. However, I am using high-dimensional FEs, which means it is not feasible to do

reg y i.x1

because the number of different x1 is too large.

How can I achieve it?

[Carlo Lazzaro]

David:
have you already taken a look at -areg-?

[Phil Bromiley]

xtreg is an other obvious approach. The results are often identical to areg, but xtreg has some different options.