-table-'s main job is to produce tables,
not resultssets;
as you don't like what -replace- does,
your main option is to reproduce
what -table- does by other means,
e.g.
egen count1 = count(var1), by(state)
egen count3 = count(var3), by(state)
gen newvar = count1 / count3
Note that, necessarily, each value
of these variables is repeated within
groups of -state-. -egen, tag()- offers
one handle here.
Alternatively, -egen-haters can drill
one level down and do it all as an
exercise in -by:-, _n and _N.
Note that Roger Newson's utilities
do not offer a solution here, as they
are firmly wedded to a data reduction
approach.
Nick
[email protected]
Rijo John
> I am using a simple table command
>
> table state, c(n var1 n var2 n var3) row
>
> Now I want to use the result of this table for generating new
> variables.
>
> Say, gen newvar = N(var1)/N(var3). One way is using the
> "replace" option
> with the table command. But that cuses the data in memory to
> lose. And I
> will have to insheet the data set again. With out that
> option is there a way out?
*
* 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/
