Re: st: RE: graph interaction curve

[Jorge Eduardo Pérez Pérez <perez.jorge@ur.edu.co>]

I think the commands -xi3- and -postgr3- (which can be installed
writing -net install spost9_ado.pkg-) can achieve this, as shown in
the following example, although I'm not sure if this is what you're
looking for

use http://www.ats.ucla.edu/stat/stata/notes/hsb2, clear
gen mathsq=math^2
* math would be X, ses is the moderator
xi3: reg science mathsq i.ses*math
postgr3 math, by(ses) asis(mathsq _Ise*)

Hope this helps,
_______________________
Jorge Eduardo Pérez Pérez


On Fri, Feb 25, 2011 at 11:27 AM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
> The short answer is No. There isn't an easy way, but there is a way to solve this nevertheless. Maarten already posted a concrete example at
>
> http://www.stata.com/statalist/archive/2011-02/msg01240.html
>
> That is, get a series of predictions for different values of the moderator and plot them as a series of lines in the (X, Y) space.
>
> So, the same question gets the same answer.
>
> Otherwise we need to see something new from you, e.g. specific detail on how you wrote code that doesn't work, etc.
>
> Nick
> n.j.cox@durham.ac.uk
>
> Fabio Zona
>
> I am really having hard time to graph an interacion between a curve and a linear moderator.
>
> this is the equation:
>
> Y = a + X +  X^2  + moderator  +  X * moderator
>
> Is there an easy way to graph the curve and see how it changes, for different values of the moderator?
>
> Also, I have another equation (more complex) with the same challenge:
> Y = a + X +  X^2  + moderator  +  X*moderator  + X^2 * moderator
>
>
> *
> *   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/
>

*
*   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/