Juanita,
I could be wrong, but I think reshape allows you to specify where the
'j' portion of the data resides within the variable name using the '@'
symbol.
so, with data as you've described :
> date code1_varx code1_vary code2_varx code2_vary
> 1990 10 20 10 30
> 1991 10 20 10 30
you could type
. reshape long code@_varx code@_vary, i(id) j(code)
which should give you
> date code_varx code_vary code
> 1990 10 20 1
> 1990 10 30 2
> 1991 10 20 1
> 1991 10 30 2
then you would just have to rename each code_varx to varx to vary, as follows:
unab codevars : code_*
foreach V of local codevars {
local rename=subinstr("`V'","code_","",.)
rename `V' `rename'
}
Jacki
On 10/4/06, Riano, Juanita <[email protected]> wrote:
Philipp,
Thank you! your code is great but I am very sorry because I failed to
mention that my codes are six-digits random numbers,
They look like code741486 code904195
Juanita
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Philipp Rehm
Sent: Wednesday, October 04, 2006 12:08 PM
To: [email protected]
Subject: Re: st: RE: RE: Problem with "reshape"
This may work:
forvalues i = 1(1)300 {
foreach XY in x y {
rename code`i'_var`XY' var`XY'_`i'
}
}
reshape long varx_ vary_, i(date) j(code)
rename varx_ varx
rename vary_ vary
You may need to make use of capture, in case your individuals are not
consecutievely numbered.
HTH,
Philipp
Riano, Juanita wrote:
> Thanks Justin, but 'stack' doesn't work because I need to preserve the
> individuals' code.
>
> Juanita
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of White,
> Justin
> Sent: Wednesday, October 04, 2006 11:47 AM
> To: [email protected]
> Subject: st: RE: Problem with "reshape"
>
> Try the 'stack' command
>
>
>
> Justin White
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]] On Behalf Of Riano,
> Juanita
> Sent: Wednesday, October 04, 2006 11:43 AM
> To: [email protected]
> Subject: st: Problem with "reshape"
>
> Dear all,
>
> I have a wide dataset organized as follows
>
> date code1_varx code1_vary code2_varx code2_vary
> 1990 10 20 10 30
> 1991 10 20 10 30
> ...
>
> Where: code1- code300 identify individuals and varx vary are the
> varnames (I have 9 variables).
>
> What I want is a long dataset that looks like this
>
> date code varx vary
> 1990 1 10 20
> 1991 1 10 20
> ..
> 1990 2 10 30
> 1991 2 10 30
> ..
>
> Using the reshape command hasn't worked for me so far because the stub
> that identifies the name of the variable is at the end of it:
> no xij variables found r(111);
>
> I also tried to transpose the data using the "xpose" command to
> manipulate the names but when you use "xpose" twice the string
> variables are dropped,,,
>
> I would really appreciate any suggestion on how to proceed,,,
>
> Thank you very much,
>
> Juanita Riano
>
>
>
>
> *
> * 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/
>
> *
> * 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/
*
* 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/
