-egen, tag()- is an automated way of getting Antoine's -ok- variable. (His method is certainly OK.)
Nick
n.j.cox@durham.ac.uk
Antoine Terracol
you could generate the means and then plot them
sysuse auto, clear
bysort rep78 : egen m_mpg=mean(mpg)
bysort rep78 : egen m_weight=mean(weight)
twoway scatter m_mpg m_weight
here each observation contribute to the plot, which could thus take time
to draw. You could limit yourself to one obs per group using
bysort rep78 : gen ok=(_n==1)
and then
twoway scatter m_mpg m_weight if ok
On 31/03/2010 13:55, Fernando Terrés wrote:
> Martin, perhaps I didn't explain it correctly
> Let us say that, in the auto.dta dataset, I want to plot the mean values
> of mileage against the mean values of weight, both means grouped by
> repair record.
> I could do:
>
> *************
> sysuse auto, clear
> collapse mpg weight, by(rep78)
> twoway (scatter mpg weight)
> *************
>
> My problem is that I need to do the same task for several categorical
> variables of a large file, and every collapse forces me to upload the
> dataset again.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
