Though this does not exactly answer your question, could you jitter
the x-axis and use three different colors?  Even though you are
calculating the probabilities at the same points, shifting them just
a bit left and right makes it possible to plot all three so long as
you do not employ really fat lines.

RWW

On Oct 23, 2006, at 10:41 AM, Jim Battista wrote:

> So I have a multinomial logit model and want to display my results
> graphically.
>
> I can do this easily enough without confidence intervals using the
> xpost
> tools after the estimation.  Or I can use the spost tools inside Stata
> to generate confidence intervals, and use rarea to plot them.
>
> Now, I'm not ashamed of my confidence intervals.  They're beautiful
> confidence intervals that I'm happy to show.  But, sometimes, some of
> the confidence intervals overlap, which makes my existing figures
> unsatisfactory.  I've found the drarea package for Stata which shades
> overlapped areas as the average of the two areas, but that only
> lets you
> throw up two CI's, whereas I have three, and you can't add to it with
> addplot().
>
> Dumping the data that Stata uses to create the graphs is trivial.
> So is
> there another graphing / charting package that will let me overlay one
> CI on another with the top one such that I can still see the other one
> underneath it?  That will let me set a transparency value for the top
> one, or that will average or sum the colors / grey values for
> overlapped
> areas without me having to manually calculate the areas to shade them
> directly?  What I'd really like is something such that a light grey
> overlapping a medium gray turns into a dark (=light+medium) grey.
>
> Apologies if this is trivially easy in R.  I've looked, but I am not
> always the best searcher.
>
> Jim
>
> --
> James S. Coleman Battista
> Dept. of Political Science, Univ. of North Texas
> [log in to unmask]    (940)565-4960
>
> **********************************************************
>             Political Methodology E-Mail List
>        Editor: Karen Long Jusko <[log in to unmask]>
> **********************************************************
>        Send messages to [log in to unmask]
>  To join the list, cancel your subscription, or modify
>           your subscription settings visit:
>
>          http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************

Robert W. Walker
Assistant Professor
Department of Political Science
Program in Applied Statistics and Computation
Washington University in Saint Louis
Campus Box 1063
One Brookings Drive
Saint Louis, Missouri 63130-3899
rww at wustl.edu
http://rww.wustl.edu

**********************************************************
             Political Methodology E-Mail List
        Editor: Karen Long Jusko <[log in to unmask]>
**********************************************************
        Send messages to [log in to unmask]
  To join the list, cancel your subscription, or modify
           your subscription settings visit:

          http://polmeth.wustl.edu/polmeth.php

**********************************************************