Herv� wrote:
> I refer to a problem I mentioned earlier and which was solved
> by the second solution you proposed originally (the one with
> the program prsign, eclass): in my estout output, I want to
> have the results of two regressions:
>
> regress y var1 var2
> regress y var1 var2 var3.
>
> In the predicted signs column, I need to have the three
> variables var1, var2 and var3, although >the first model only
> uses var1 and var2.
>
> It seems to me that your latest solution (with tempname b) is
> related to only one model.
No, not necessarily. Consider:
sysuse auto
regress price mpg weight
est sto model0
regress price mpg weight foreign
estimates store model1
tempname b
matrix `b' = e(b)
matrix `b'[1,1] = -999 //neg. sign for mpg
matrix `b'[1,2] = +999 //pos. sign for weight
matrix `b'[1,3] = .z //no sign for foreign
matrix `b'[1,4] = 998 //? for _cons
eret2 matrix signs = `b'
estimates store pred
estout pred model0 model1, cells("signs(pat(1 0 0)) b(pat(0))") ///
substitute(-999 - 999 + 998 ?)
ben
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
