Michael Blasnik <[email protected]> has found a problem with the
-normal- option of -histogram- when used with the -by()- option:
> I just tried using histogram with the by option and requesting a normal
> density curve along the lines of:
>
> histogram myvar if this=1, normal by(groupvar)
>
> The resulting graphs for each groupvar have normal density curves that
> apparently differ from graph to graph, but they are mostly wrong -- they do
> not come close to the curves drawn when issuing the same command on
> individual groups and they often seem to have nothing to do with the data in
> the plot. Is this a bug or am I using this incorrectly?
It is definitely a bug. Looking at the code, the normal densities are all the
same, and are representative of the entire data (within -if- and -in-
conditions) without regard to the -by()- groups. Basically, the -normal-
option ignores the -by()- option.
I will make it a priority to fix this soon.
--Jeff
[email protected]
*
* 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/
