Second try:
levelsof octy, local(D)
gen tshare = .
qui foreach d of local D {
su tradeshare if octy == "`D'", meanonly
replace tshare = r(mean) if dcty == "`D'"
}
bysort octy : egen x1 = total(tshare * dist)
Nick
[email protected]
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Nick Cox
> Sent: 28 February 2007 22:46
> To: [email protected]
> Subject: st: RE: Data management
>
>
> Looks like
>
> bysort octy : egen x1 = total(tradeshare * dist)
>
> Nick
> [email protected]
>
> Roy,Suryadipta
>
> > I had posted a similar query sometime back and did not get
> > any response- so probably my question was not very clear.
> > Here I have modified the query and it would be extremely
> > helpful to get some help. The dataset is as follows:
> >
> > octy dcty dist tradeshare x1
> > a b 10 0.13 12.95
> > a c 15 0.13 12.95
> > a d 25 0.13 12.95
> > b a 10 0.44 13.15
> > b c 30 0.44 13.15
> > b d 25 0.44 13.15
> > c a 15 0.22 16.2
> > c b 30 0.22 16.2
> > c d 5 0.22 16.2
> > d a 25 0.21 15.35
> > d b 25 0.21 15.35
> > d c 5 0.21 15.35
> >
> > Value of x1 for octy a (=12.95) has been generated by the
> > formula:
> > tradeshare(b)*dist(a,b)+tradeshare(c)*dist(a,c)+tradeshare(d)*
> > dist(a,d).
> >
> > Similarly, x1 for octy b
> > (=13.15)=tradeshare(a)*dist(b,a)+tradeshare(c)*dist(b,c)+trade
> > share(d)*dist(b,d),
> > and so on for octy c and d.
> >
> > I can do the above for 4 countries in Excel but with more
> > than 150 countries, it gets extremely difficult. Any
> > cooperation is highly appreciated.
*
* 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/
