Richard,
Many thanks for your suggestion. I tried using globals, but again it didn't
seem to work. I'm attaching part of the code I'm using (it'll hopefully be
'readable'). The strange thing is that I'm just having problems with the
outreg line of the program (STATA doesn't recognise the loop `var' in the
name of the out file...).
Thanks again for help and ideas!
Andrea
global varlist "text wood paper chem nonmet bamet metal"
global xvar "adj ldist eec9mxc ibemxc eftmxc grceec9mc grceec9xc ibeeec9mc
ibeeec9xc ibegrcmc ibegrcxc efteec9mc efteec9xc eftgrcmc eftgrcxc eftibemc
eftibexc eec9nmmc eec9nmxc grcnmmc grcnmxc ibenmmc ibenmxc eftnmmc eftnmxc
lingtie"
global xvarsetPRE "adjsePRE ldistsePRE eec9sePRE ibesePRE eftsePRE
grceec9msePRE grceec9xsePRE ibeeec9msePRE ibeeec9xsePRE ibegrcmsePRE
ibegrcxsePRE efteec9msePRE efteec9xsePRE eftgrcmsePRE eftgrcxsePRE
eftibemsePRE eftibexsePRE eec9nmmsePRE eec9nmxsePRE grcnmmsePRE grcnmxsePRE
ibenmmsePRE ibenmxsePRE eftnmmsePRE eftnmxsePRE lingtiesePRE"
global xvarsetPOS "adjsePOS ldistsePOS eec9sePOS ibesePOS eftsePOS
grceec9msePOS grceec9xsePOS ibeeec9msePOS ibeeec9xsePOS ibegrcmsePOS
ibegrcxsePOS efteec9msePOS efteec9xsePOS eftgrcmsePOS eftgrcxsePOS
eftibemsePOS eftibexsePOS eec9nmmsePOS eec9nmxsePOS grcnmmsePOS grcnmxsePOS
ibenmmsePOS ibenmxsePOS eftnmmsePOS eftnmxsePOS lingtiesePOS"
capture program drop regse
program regse
local dep `1'
local xvar "`2'"
local cons `3'
regress `dep' `xvar' `cons', robust nocons
predict resid, r
log on
sktest resid
jb1 resid
log off
drop resid
*need to re-estimate the R-squared here!
capture drop yhat
capture drop r2
capture drop constant
gen constant=1
mkmat $xvar constant, mat(xvar)
matrix b = e(b)'
matrix yhat = xvar * b
svmat yhat, name(yhat)
quietly corr `dep' yhat
gen r2 = r(rho)^2
di ":::::::::: R-squared using method 2 :::::::::::"
di "R2 = " r2
outreg using PREPOS`var'ss, bdec(3) coefastr se 3aster nor2
addstat("R2",r2,"Normality",$jb) adec(4,2) append
end
foreach var of global varlist {
log using chow2stage`var'ss.log, replace
log off
xtreg srem`var' srep`var' srec`var' lrgdp95pcm lrgdp95pcx lbrer dbt`var'
d79-d87 if year<1988, fe
outreg using PREPOS`var'ss, bdec(3) coefastr se 3aster nor2 replace
xtreg srem`var' srep`var' srec`var' lrgdp95pcm lrgdp95pcx lbrer dbt`var'
d93-d99 if year>=1992, fe
outreg using PREPOS`var'ss, bdec(3) coefastr se 3aster nor2 append
bysort mx: egen msrem`var'PRE=mean(srem`var') if year<1988
by mx: egen msrep`var'PRE=mean(srep`var') if year<1988
by mx: egen msrec`var'PRE=mean(srec`var') if year<1988
by mx: egen mdbt`var'PRE=mean(dbt`var') if year<1988
by mx: egen msrem`var'=mean(srem`var') if year>=1992
by mx: egen msrem`var'POS=max(msrem`var')
by mx: egen msrep`var'=mean(srep`var') if year>=1992
by mx: egen msrep`var'POS=max(msrep`var')
by mx: egen msrec`var'=mean(srec`var') if year>=1992
by mx: egen msrec`var'POS=max(msrec`var')
by mx: egen mdbt`var'=mean(dbt`var') if year>=1992
by mx: egen mdbt`var'POS=max(mdbt`var')
regse uset`var'PRE "$xvarsetPRE" conssePRE
matrix rssPREset=e(rss)
matrix nPREset=e(N)
regse uset`var'POS "$xvarsetPOS" conssePOS
matrix rssPOSset=e(rss)
matrix nPOSset=e(N)
save chowfetse`var'ss, replace
log close
}
----- Original Message -----
From: "Richard Williams" <[email protected]>
To: <[email protected]>
Sent: Wednesday, August 17, 2005 3:45 PM
Subject: Re: st: Problems with program
> At 03:39 PM 8/17/2005 +0100, Andrea Molinari wrote:
> >Dear all,
> >
> >I am trying to build a program to estimate the same procedure but with
> >seven (slightly) different sets of data. For this, I'm using the
forvalues
> >loop with seven locals, but this stops if there is a program (which also
> >uses those seven sets of data) inside the loop.
> >
> >I tried putting the programs outside the loop, and 'call' it inside the
> >loop, but in that case the program does not recognise the locals of the
> >loop.
> >
> >Does anyone have any suggestions for solving this problem?
>
> It would help to see some of your code, but one possibility might be to
use
> global variables instead of local variables.
>
> -------------------------------------------
> Richard Williams, Notre Dame Dept of Sociology
> OFFICE: (574)631-6668, (574)631-6463
> FAX: (574)288-4373
> HOME: (574)289-5227
> EMAIL: [email protected]
> WWW (personal): http://www.nd.edu/~rwilliam
> WWW (department): http://www.nd.edu/~soc
>
> *
> * 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/
>
*
* 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/
