Alejandro Ria�o
> > I have the following monthly dataset:
> >
> > date Argentina Brazil Colombia .....
> > 01/01/90 x1 y1 z1
> > 02/01/90 x2 y2 z2
> > . . . .
> > . . . .
> >
> > and so on..
> >
> > where Xi, Yi and Zi are values for a given time-series
> >
> > I'd like this change this dataset into a long format,
> > like this:
> >
> >
> > date Country Serie
> > 01/01/90 ARG x1
> > 02/01/90 ARG x2
> > 01/01/90 BRA y1
> > 02/01/90 BRA y2
> > 01/01/90 COL z1
> > 02/01/90 COL z2
Radu Ban
> first you have to rename your countries so that you have a
> common stub
> "serie"
>
> for example you could do (i just learned to this recently on
> Statalist, namely from Nick Cox):
>
> local fullnames "Argentina Brazil Colombia..."
> local codes "ARG BRA COL..."
>
> local i = 1
> foreach x of local fullnames {
> local y : word `i' of `codes'
> rename `x' serie`y'
> local i = `i' + 1
> }
>
> *now move on to actual reshaping
>
> reshape long serie, i(date) j(country) string
>
> *note that you have to add the [string] option at the end
> *since the suffix is in string format
>
> hope this works, i didnt have time to test it.
As Radu says, the key is getting a common stub. This
is emphasised rather more strongly at
http://www.stata.com/support/faqs/data/reshape3.html
than at [R] reshape.
If the codes "ARG" "BRA" etc. were not crucial,
another way to do it would be something like:
foreach v of var Argentina-Zimbabwe {
rename `v' Serie`v'
}
reshape long Serie, i(date) j(country) string
Nick
[email protected]
*
* 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/
