st: Number of observations reported by xtreg with fixed effects

[Daniel Brown <dzbrown2010@gmail.com>]

Dear Statalisters,


I am confused with the number of observations reported by xtreg with
the fixed effects option.

I have a panel with 3820 individuals who are in 2490 households
(hence, there are aproximately 1.5 individuals per household). I have
2 years of data (and a total of 6431 individual observations).

When I run xtreg with individual fixed effects, the number of
observations reported is 6263 (and number of groups is 3753).

When I run xtreg with household fixed effects the number of
observations reported is 5797 (and the number of groups is 2483).

Note that I don't use both individual and household fixed effects in
the same regression, since this would not make sense.

I do not understand why the same regression report a smaller number of
observations when I use household fixed effects (compared to when I
use individual fixed effects).

Any help is greatly appreciated!
Thank you very much,
Daniel
*
*   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/