st: xtreg , re: if var dropped, whence coef. estimates from within and between for error estimates in RE?‏

[redacted user <redacted>]

Hello, All,

The XT manual [Stata 10, p. 411] indicates that for xtreg, re, the idiosyncratic error component, sigma_e, is estimated from the residuals of a within estimate.

I would like to know whether I can safely assume that when time invariant variables are present (and would be dropped if using xtreg, fe directly to get within estimates), that xtreg, re uses the estimates (for the above-mentioned idiosyncratic error component) from a within regression that simply drops such variables. A similar question applies to estimating sigma^2_u and dropped variables in a between regression (the estimates of which xtreg, re uses to compute sigma^2_u).

Thank you.

redacted
_________________________________________________________________
Bing™ brings you maps, menus, and reviews organized in one place. Try it now.
http://www.bing.com/search?q=restaurants&form=MLOGEN&publ=WLHMTAG&crea=TXT_MLOGEN_Local_Local_Restaurants_1x1
*
*   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/