Sounds as if you need something like
line response time , by(phase) || lfit response time
If this is wrong, and you don't get a better answer, show us some
realistic data.
A concrete example usually beats a word description.
Nick
On Sun, Mar 24, 2013 at 8:34 PM, Dan Brossart <brossart@tamu.edu> wrote:
> I am trying to generate a graph for single-case data (20 time points total)
> where the baseline has 10 occasions of measurement and the treatment phase
> has 10 occasions of measurement. I need to be able to graph the baseline and
> treatment as a single graph with a vertical line separating the two phases.
> In each phase I need to have a line connecting each data point (but not from
> phase 1 to phase 2). I also need the ability to add a regression line to
> each phase - each phase gets its own line, independent of the other phase
> regression line.
>
> I've looked through Mitchell's "A Visual Guide to Stata Graphics" 3rd Ed.,
> but there isn't an example like this and so far I haven't found anything
> like this in the archives. My efforts to use twoway have so far failed to
> produce anything close. Using twoway with xrescale and a by statement gives
> me two separate graphs and I really want to do this in a single graph.
> Ideas?
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
