Degrees of freedom

[Maximilian Dekorsy]

Hello!
I am wondering how I might calculate degrees of freedom using an fixed-effects regression model.

Thanks for your advice.
Best regards,
Max

[Carlo Lazzaro]

Maximilian:
assuming that you're referring to a panel data analysis with fixed-effect specification, please find below a Stata example to retrieve what you're after:

use http://www.stata-press.com/data/r13/nlswork 
xtreg ln_w grade age c.age#c.age ttl_exp c.ttl_exp#c.ttl_exp tenure c.tenure#c.tenure 2.race not_smsa south, fe
ereturn list

For model degrees of freedom, take a look at: e(df m);
for residual degrees of freedom, take a look at: e(df r);
for degrees of freedom for absorbed effect, take a look at: e(df a);
for numerator degrees of freedom for F statistic, take a look at: e(df b).

[Lise Pe]

Mr Lazzaro,
Thank you for your help.
Would you have an advice for calculating degree of freedom in models for which it is not automatic in Stata (and for which the formula cannout be found in the literature) ?
More specifically, for the FMOLS/DOLS/CCR models (with the xtcointreg command) ?
Regards

[Carlo Lazzaro]

Lise:
I'm not familiar with those estimators.
Sorry for that.