Well, you could put all those commands in one long command like so:
#delimit;
forvalues i=1 to 16 {;
g jobstatus`i'= 86 *(jobnum`i'== 003)+ 77*(jobnum`i'== 004)
+...+
8 *(jobnum`i'== 909);
};
but that is still an awful lot of typing. Another way is make a
dataset of 500obs with just the two codes, and -merge- this on to your
original data, after applying -reshape- to your original data. But
that is also a lot of work.
I like this approach (it seems more easily extensible and more
transparent, somehow):
#delimit;
la def jobstat 86 "003";
la def jobstat 77 "004", modify;
...
la def jobstat 8 "909", modify;
tempname t;
forvalues i=1 to 16 {;
la val jobnum`i' jobstat;
decode jobnum`i' , g(`t');
g jobstatus`i'=real(`t');
drop `t';
};
On 2/23/06, Marino, Jennifer <[email protected]> wrote:
> I am doing the following and fervently believe that there are more
> elegant, faster ways to do this. (I am working in Stata 8.2 but will be
> updating to 9 any day now.) Occupational history codes jobnum1-jobnum16
> each take one of 500 values from 003 to 909, and I am assigning a
> Nam-Powers status value (1 to 100) to each occupation held.
>
> forvalues i=1 to 16{
> gen jobstatus`i'=.
> }
>
> forvalues i=1 to 16{
> replace jobstatus`i'= 86 if jobnum`i'== 003
> }
> forvalues i=1 to 16{
> replace jobstatus`i'= 77 if jobnum`i'== 004
> }
> ...
> forvalues i=1 to 16{
> replace jobstatus`i'= 08 if jobnum`i'== 909
> }
>
> It's not a problem to me to do this in this kludgy fashion, but I'd like
> to learn how Stata experts would do it. Please enlighten me. Thank you
> very much.
*
* 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/
