Marcello mentions -eqprhistogram-, which is a program on SSC.
A search using -findit- reveals nothing. Various people
have played with this in this past, but none that I know
of got as far as a truly general program which they made
public.
This is (now) either fairly easy or a little more difficult. The
easiest case is when your bins are such that there are
no bins other than those beyond the extremes with zero
frequencies. The more difficult case is when you need
to cope with zero bins in the middle of the distribution.
The "(now)" qualifier reflects the existence of -twoway__histogram_gen-
discussed by David Harrison in SJ 5(2) 2005.
The following shows some technique but falls far short of
a general recipe.
sysuse auto, clear
twoway__histogram_gen mpg, start(0) width(2) freq gen(mpgfreq mpgbin)
levelsof mpgbin, local(have)
numlist "11(2)43"
local should "`r(numlist)'"
local need : list should - have
local i = _N
qui foreach v of local need {
replace mpgbin = `v' in `i'
replace mpgfreq = 0 in `i'
local --i
}
twoway line mpgfreq mpgbin, sort
Nick
[email protected]
Marcello Pagano
> Do we have a frequency polygon in Stata? On top of eqprhistogram?
*
* 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/
