-ciplot- is a user-written program. It doesn't work
by any kind of reduction or collapse, so there isn't
a simple mapping between its data structure and that
of the rest of the dataset that is also exploitable
in this way. Hence while what you say is not
always true about -ciplot-, it really doesn't lend
itself to what you want.
The alternatives seem to include your rewriting -ciplot- so
that something like this is wired in; or, as you
say, using -eclplot-.
Nick
[email protected]
Daniel Egan
>
> A minor detail of the options with -ciplot-: I would like to
> be able to use
> labels with the mean markers that represent various things (right now,
> number of observations in each category).
>
> sysuse auto, clear
> egen n=count(foreign), by(foreign)
> table foreign n
> ciplot mpg, by(foreign) mlab(n)
>
> These labels always only correspond to the value of the first
> categorical
> label, i.e., category 2's label is the value for category one.
>
> How might I go about having each label value correspond to
> the category
> specific value? Would I need eclplot for this?
*
* 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/
