How about something like this:
sum daysupply , meanonly
forvalues daynum = 2/`r(max)' {
replace date`daynum' = date1 + `daynum'-1 if daysupply>=`daynum'
}
The first line summarizes daysupply, which puts the maximum value of
daysupply in `r(max)'. (The 'meanonly' option suppresses the output of the
sum command, as well as speeding it up a bit by not calculating the
variance of daysupply).
Then you loop through the values of daynum that you need (i.e., from 2
through the maximum value of daysupply. The -if- condition in the replace
command ensures that the command is only executed for the appropriate
observations.
(If you knew beforehand that the maximum value of daysupply was 10, as in
your example, you could code thta directly:
forvalues daynum = 2/10 {
replace date`daynum' = date1 + `daynum'-1 if daysupply>=`daynum'
}
but that would be less general, and would fail if you ran it again later
with a dataset that had larger values of daynum.
--Nick Winter
At 11:33 AM 6/8/2005, you wrote:
I'm trying to change several variables in each observation. The problem is
that the number of variables I want to update is different for each
observation and this number is a variable in that observation. I've tried
using a forvalues loop, which works if I specify a stop and start number,
but I cannot find a way to put a variable in the numlist allowing this to
change for each observation.
Here's the command that works:
forvalues daynum = 2/10 {
replace date`daynum'=date1+`daynum'-1
}
But when I replace the 10 with the variable that tells me how many date
variables to change, I get a syntax error.
forvalues daynum = 2/daysupply {
replace date`daynum'=date1+`daynum'-1
}
Any suggestions?
Thanks,
_Shawn
--------------------------------------------------------
Nicholas Winter 607.255.8819 t
Assistant Professor 607.255.4530 f
Department of Government [email protected] e
308 White Hall falcon.arts.cornell.edu/nw53 w
Cornell University
Ithaca, NY 14853-4601
*
* 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/
