Suppose you work out that the company in
question is the 42nd observation.
Then the option could be
yline(`=ret2001Q2[42]')
Suppose you are cycling over several
variables. You can still use the same idea:
foreach v of var <whatever> {
quantile `v', recast(line) rlopts(lwidth(medthick)) ///
ylabel(-.06 (.01) .06, ang(h)) yline(`=`v'[42]') ///
xlabel(0 .1 .2 .25 .3 .4 .5 .6 .7 .75 .8 .9 1) ///
xline(0 .1 .2 .25 .3 .4 .5 .6 .7 .75 .8 .9 1, lcolor(bluishgray))
}
What you are relying on here is that -quantile- does not
(permanently) affect the sort order of the data.
Nick
n.j.cox@durham.ac.uk
Eric G. Wruck
> I am generating a series of quantile plots of quarterly returns but
> would like to flag one company (out of 150 or so that comprise the
> data). I found one way of doing this, using yline after manually
> looking up that company's value for the quarter in question. For
> example,
>
> quantile ret2001Q2, recast(line) rlopts(lwidth(medthick)) ylabel(-.06
> (.01) .06, angle(horizontal)) yline(0.011) xlabel(0 .1 .2 .25 .3 .4
> .5 .6 .7 .75 .8 .9 1) xline(0 .1 .2 .25 .3 .4 .5 .6 .7 .75 .8 .9 1,
> lcolor(bluishgray))
>
> Here I looked up the 2001, 2nd quarter return for the company in
> question & saw that its return was 0.011 -- hence, yline(0.011).
>
> My question is: is there a way to pass something to yline from the
> data so that this doesn't have to be a manual operation (& hence
> susceptible to error)?
*
* 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/