Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Plotting regression constants |
Date | Mon, 8 Aug 2011 08:12:46 +0100 |
Your error report suggests some problem with producing a dataset of constants, rather than a limitation of -graph box-. I'd guess wildly that -stripplot- (SSC) would be more useful for you, but there are many graphics commands to show distributions. Nick On Mon, Aug 8, 2011 at 5:37 AM, Jennifer M Keister <jkeister@ucsd.edu> wrote: > Nick, > > The graph I get when I run the syntax indicated produces only one horizontal line (at the value of the constant), and a dot in the middle of the line. The y-axis of the graph has tickmarks at the values for the upper and lower quartiles. I would like something that more clearly indicates the standard errors. > > Since I am using the constants as (effectively) predicted probabilities, indicating the variation around the constant will be particularly helpful once I can figure out how to display multiple constants in a single graph, since that will visually display which predicted probabilities overlap with one another and which do not. * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/