It is not clear to me what you mean by "the coefficients and variables
that do vary at the
> "mother_rc" - level"
Some previous discussion of this issue can be found at:
http://www.stata.com/statalist/archive/2006-07/msg00535.html
http://www.stata.com/statalist/archive/2002-07/msg00311.html
http://www.stata.com/statalist/archive/2002-07/msg00313.html
The example below show the changes depending on the command (use
-findit xtivreg2- to locate and download):
webuse grunfeld,clear
drop if year <1954 & com<6
qui {
areg invest mval kst, ab(com)
estimates store areg_full
areg invest mval kst if com >5, ab(com)
estimates store areg_restrict
xtivreg2 inve mva kstock, fe
estimates store xtivreg2
xtivreg2 inve mva kstock, fe small
estimates store xtivreg2_small
}
estimates table areg_full areg_restrict xtivreg2 xtivreg2_small, se
stats(N r2_a) modelwidth(15)
. estimates table areg_full areg_restrict xtivreg2 xtivreg2_small, se
stats(N r2_a) modelwidth(15)
--------------------------------------------------------------------------------------
Variable | areg_full areg_restrict xtivreg2
xtivreg2_small
-------------+------------------------------------------------------------------------
mvalue | .08268515 .08268515 .08268515
.08268515
| .007895 .007895 .00781145
.007895
kstock | .11446319 .11446319 .11446319
.11446319
| .01394182 .01394182 .01379428
.01394182
_cons | 1.0301229 -7.5449016
| 3.2766423 2.6484442
-------------+------------------------------------------------------------------------
N | 105 100 100
100
r2_a | .99586829 .87503794 .77143666
.77143666
--------------------------------------------------------------------------------------
legend: b/se
Scott
On Thu, Sep 17, 2009 at 4:45 PM, Sue <[email protected]> wrote:
> So those observations are not used to estimate the coefficient of
> variables that don't change within "mother_rc", but for all others,
> including the coefficients and variables that do vary at the
> "mother_rc" - level and that's why the number of observations doesn't
> change?
>
*
* 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/
