> I want to solve this problem and get the ordering of the variables
> correct because my goal is to get a final product like the following
> table after a little work by hand:
>
> y early_y late_y
> X1 $ $
> X2 $ $ $
> X3 $ $ $
> X4 $
I read the rest of the posting, and this can be done by applying
-replace- before each regression.
gen x1=.
gen x2=.
gen x3=.
replace x1=current_x1
replace x2=current_x2
replace x3=current_x3
reg y x1-x3
outreg2 using stuff
replace x1=early_x1
replace x2=early_x2
replace x3=early_x3
reg y x1-x3
outreg2 using stuff
replace x1=late_x1
replace x2=late_x2
replace x3=late_x3
reg y x1-x3
outreg2 using stuff
Someone asked me earlier if this can be done automatically from inside
outreg2. I guess it can be done (some macro replacements), but the
syntax will get ugly, i.e. having to specify every match, which didn't
seem worth the trouble.
Roy
_________________________________________________________________
Insert movie times and more without leaving Hotmail®.
http://windowslive.com/Tutorial/Hotmail/QuickAdd?ocid=TXT_TAGLM_WL_HM_Tutorial_QuickAdd_062009
*
* 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/
