Ah! Yes indeed, that's different.
A principle often mentioned on this list, and somehow I feel a
Stata Journal tip coming on for SJ 6(2) 2006, is that when
the way ahead with -graph bar- is blocked, you get out of that
and roll your sleeves up for a second try with -twoway bar-.
Typically some preliminary spadework is needed with -gen-
and -egen-.
So,
sysuse auto
egen meanprice = mean(price), by(headroom)
egen tag = tag(headroom)
egen perc = sum(1), by(headroom)
count if headroom < .
replace perc = 100 * perc / r(N)
gen percent = string(perc, "%5.2f")
twoway bar meanprice headroom if tag, barw(0.25) ||
scatter meanprice headroom if tag, ms(none) mla(percent) mlabpos(12) legend(off)
Nick
[email protected]
Esther Kessler
> thank you for your answer. I know the command catplot, but I
> think I can't solve my problem with it, because the variable
> price is relevant for me. I would like to show the relation
> between price and headroom. In addition I would like to have
> the frequencies of the variable headroom as a label above the
> bars (price). I would like to show how often the headroom
> 1.5, headroom 2 etc. occurs in my data. All these information
> I would like to present in one graph.
>
> In my original data I have interval of wages (x-axis) and
> effort (y-axis) and I would like to show how many people are
> in each unit (wage- interval), so these units sum up to 100%.
> > I would like to have the frequencies in percent above the
> > bars as a label. I couldn't find anything in the visual guide
> > to stata graphics.
> >
> > To illustrate my problem:
> > sysuse auto
> > tab headroom
> > graph bar price, over(headroom) blabel(bar, format(5%5.2d))
> >
> > Instead of the price I would like to have 5.14 above the
> > first bar (headroom 1.5), 17.57 above the second bar
> > (headroom 2.0) and so on.
> >
*
* 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/
