Re: st: Residuals in Panel Data regression

["D. Demetriou" <dd358@cam.ac.uk>]

Prakash,

Thank you for getting back to me!


I am using Fixed-effects regression models (xtreg, fe) and I just want to 
know what properties do my regression residuals need to have in order for 
me to be able to use reliably t-statistics (for significant coefficients), 
F-statistics for joint significance, etc. for example: is normality 
important ? ( i think yes since t-statistics are designed for normal data). 
And so on...

The key question here is: do the conditions/asumptions (regarding error 
process) differ between OLS and FE??

Yours,

DD




Dimitri, it would be nice if you can restructure your question as in
what do you mean by it and what do you want to do.



Prakash

On Sat, Apr 21, 2012 at 3:28 AM, D. Demetriou <dd358@cam.ac.uk> wrote:
> Hi everyone,
>
> Hope my email finds you all well.
>
> Just a quick question here: What requirements should one have on the
> residuals of panel data regressions in order to be able to use reliably
> t-statistics etc.
>
> Do these differ if one uses fixed-effects regression rather than POLS?
>
> Thank you all in advance!
>
> Yours,
>
> Dimitri
> *
> *   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/

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