Re: st: how to get a two-line-graph for predicted probabilites in logistic regression model with a metric controll variable (SE10)

[Maarten Buis <maartenlbuis@gmail.com>]

For older versions of Stata you can use the -adjust- command for that like so:

*------------------ begin example ------------------
sysuse nlsw88, clear
xi: logit married age i.south*i.union

preserve
adjust age, by(south union) ci replace pr
twoway rarea lb ub south if union == 0, astyle(ci) ||    ///
       rarea lb ub south if union == 1, astyle(ci2) ||   ///
       line pr south if union == 0 ||                    ///
       line pr south if union == 1,                      ///
       legend(order(1 "95% CI" 3 "union" 4 "non-union")) ///
       scheme(s2color) ylab(,angle(horizontal))          ///
       xlab(0 "North" 1 "South")                         ///
       xtitle(area of residence)                         ///
       ytitle(predicted probability of being married)
restore
*------------------- end example -------------------
(For more on examples I sent to the Statalist see:
http://www.maartenbuis.nl/example_faq )

Hope this helps,
Maarten

On Tue, May 28, 2013 at 3:53 PM, Prof. Dr. Claudia Schuchart wrote:
> I did a graph with SE10 for predicted probabilites in a logistic regression
> model. I controlled for intelligence (metric), and considered two indendent
> binary vaiables as well as the interaction between them (N=187). I want to
> have a two-line-graph (holding intelligence constant), instead I got
> multiple graphs for any intelligence value:
>
> xi: logit y i.a*i.b IQ
> predict y, pr
> separate y, by(b)
> graph twoway line y0 y1 a , xlabel(1 2)
>
> How can I get  a two line graph only?

---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

http://www.maartenbuis.nl
---------------------------------
*
*   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/