I can't follow this. Your main issue seems to be that
you think the results you get are wrong, or at least
not what you want. It is difficult to comment on that.
However, it certainly is possible to generate
histogram numbers without getting the histogram
itself.
. search histogram
points to (inter alia)
help twoway__histogram_gen . . . . . . . . . . . . . . Histogram subroutine
(help twoway__histogram_gen)
SJ-5-2 gr0014 . . . . . . . Stata tip 20: Generating histogram bin variables
. . . . . . . . . . . . . . . . . . . . . . . . . . . . D. A. Harrison
Q2/05 SJ 5(2):280--281 (no commands)
tip illustrating the use of twoway__histogram_gen for
creation of complex histograms and other graphs or tables
Nick
[email protected]
[email protected]
> Yes, CPI is consumer price index in this case.
>
> My data set is an unweighted panel: I have around 50 months and around
> 450 different item groups in each month in my data set. The number of
> single price quotes in each item group is useally slightly different
> between months (and between item groups). To be able to draw
> a histogram
> the data is converted to the long form.
>
> I have tried to do the calculations first. The problem is that the
> histogram (distribution) becomes way too narrow with this method. I
> could get the correct weighted histogram at the whole CPI level, if I
> could sum up the individual item level histograms.
>
> To sum up, it would enought if I just could save the frequencies in
> histograms in matrix or vector form and then export them out
> of Stata as
> data. By so far, I've been not able to see the data (with the
> exception
> of bar labels option) - just the picture. I can save the picture etc,
> but does someone know how I could save the frequency data of the
> histogram?
*
* 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/
