You can save yourself some typing by building up the -scatter- command
in a loop.
sysuse auto, clear
egen labcat = cut(price), group(10)
qui centile price
local median = r(c_1)
forvalues i = 0/9 {
qui centile price if labcat == `i'
local labsize`i' = sqrt(r(c_1)/`median')
}
#delimit ;
forvalues i = 0/9 {;
local scatter "`scatter' (scatter mpg weight if labcat==`i',
msymbol(i) mlabel(make) mlabp(0) mlabsize(*`labsize`i''))";
};
#delimit cr
twoway `scatter', legend(off)
Friedrich
On 10/19/07, Philipp Rehm <[email protected]> wrote:
> Maarten,
>
> many thanks - that's a very nice work-around!
>
> Thanks again,
> Philipp
>
> Maarten buis wrote:
> > --- Philipp Rehm <[email protected]> wrote:
> >> I would like to produce a scatter plot in which the labels are
> >> "weighted", i.e. are proportional to the size of what they indicate.
> >
> > You can control the size of a label using the -mlabsize()- option, and
> > you can use that to make the kind of graph you want. See the example
> > below. Problem with these graphs is that the labels pretty soon become
> > illegible due to small size or overplotting.
> >
> > *------------- begin example ---------------
> > sysuse auto, clear
> > egen labcat = cut(pri), group(10)
> > qui centile pri
> > local median = r(c_1)
> > forvalues i = 0 / 9 {
> > qui centile pri if labcat == `i'
> > local labsize`i' = sqrt(r(c_1)/`median')
> > }
> > twoway scatter mpg wei if labcat == 0, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize0') || /*
> > */ scatter mpg wei if labcat == 1, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize1') || /*
> > */ scatter mpg wei if labcat == 2, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize2') || /*
> > */ scatter mpg wei if labcat == 3, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize3') || /*
> > */ scatter mpg wei if labcat == 4, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize4') || /*
> > */ scatter mpg wei if labcat == 5, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize5') || /*
> > */ scatter mpg wei if labcat == 6, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize6') || /*
> > */ scatter mpg wei if labcat == 7, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize7') || /*
> > */ scatter mpg wei if labcat == 8, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize8') || /*
> > */ scatter mpg wei if labcat == 9, /*
> > */ msymbol(i) mlabel(make) /*
> > */ mlabp(0) mlabsize(*`labsize9') /*
> > */ legend(off)
> > *------------------ end example ---------------------
> > (For more on how to use examples I sent to the Statalist, see
> > http://home.fsw.vu.nl/m.buis/stata/exampleFAQ.html )
> >
> > Hope this helps,
> > Maarten
> >
> >
> > -----------------------------------------
> > Maarten L. Buis
> > Department of Social Research Methodology
> > Vrije Universiteit Amsterdam
> > Boelelaan 1081
> > 1081 HV Amsterdam
> > The Netherlands
> >
> > visiting address:
> > Buitenveldertselaan 3 (Metropolitan), room Z434
> >
> > +31 20 5986715
> >
> > http://home.fsw.vu.nl/m.buis/
> > -----------------------------------------
*
* 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/
