Re: st: RE: graph interaction curve

[Fabio Zona <fabio.zona@unibocconi.it>]

does this solution work only with the functions c. ## ??
can I calculate the interactions as products of variables (standardized variables) and the apply this method? If so, why doesn't it work with me?
f






----- Messaggio originale -----
Da: "Nick Cox" <n.j.cox@durham.ac.uk>
A: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Inviato: Venerdì, 25 febbraio 2011 17:27:18 GMT +01:00 Amsterdam/Berlino/Berna/Roma/Stoccolma/Vienna
Oggetto: st: RE: graph interaction curve

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/