Re: st: How can I make a graphic of regression results using only significant effects?

["Alvaro Zevallos" <azevallos@grade.org.pe>]

Hi Maarten!



Thanks a lot for your advice, it was very useful for me, but I didn't want 
to remove the interaction effect for the model, only for the graph. When I 
was graphing the results I saw that the lines were not parallel because of 
the presence of interaction effect in the model. It's worth noting that the 
interaction effect is not significant. so - like you mentioned in the last 
email - I think it`s interesting to focus on the statistically not 
significant interaction.. In order to do that I would need a graph with two 
parallel lines, but - postgr3 - doesn't give me an option.  Anyway, I solved 
this problem changing all the values of the interaction variable for the 
mean.



Thanks for all



Alvaro.







----- Original Message ----- 
From: "Maarten buis" <maartenbuis@yahoo.co.uk>
To: <statalist@hsphsun2.harvard.edu>
Sent: Saturday, September 16, 2006 3:28 AM
Subject: Re: st: How can I make a graphic of regression results using only 
significant effects?



> Alvaro:
> -postgr3- graphs the results of your model, and your model contains an
> interaction, so -postgr3- includes that interaction. Whether or not the
> interaction is significant isn't relevant here, you forced that
> interaction to be in the model, so it is in the model. If you don't
> want that interaction just type remove the interaction from your model,
> and than call -postgr3-. However, blindly removing variables that are
> insignificant is bad practice. You included that interaction in your
> model, so apperently you thought that that interaction is interesting.
> The fact that it proves to be insignificant is than also an interesting
> result, so than you should include it in your model to show that it is
> insignificant.
>
> HTH,
> Maarten
>
> --- Alvaro Zevallos <azevallos@grade.org.pe> wrote:
> > I am working in a model that uses an interaction between two dummies
> > variables. The command that I am using is:
> >
> > --  xi3: reg Y   i.dummy1*dummy2  X
> >
> > and the results are
> >
> > dummy1                      (Significant)
> > dummy2                      (Significant)
> > dummy1* dummy2       (Not significant)
> > X                                 (Whatever)
> >
> > Then, I use the -  postgr3 - command for make a graphic of the
> > results but Stata shows me a figure that consider the not significant
> > effect (the interaction). How can I avoid that Stata considers this
> > effect?
> -----------------------------------------
> Maarten L. Buis
> Department of Social Research Methodology
> Vrije Universiteit Amsterdam
> Boelelaan 1081
> 1081 HV Amsterdam
> The Netherlands
>
> visiting adress:
> Buitenveldertselaan 3 (Metropolitan), room Z434
>
> +31 20 5986715
>
> http://home.fsw.vu.nl/m.buis/
> -----------------------------------------
>
>
>
> ___________________________________________________________
> Try the all-new Yahoo! Mail. "The New Version is radically easier to 
> use" - The Wall Street Journal
> http://uk.docs.yahoo.com/nowyoucan.html
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
>
> --
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.1.405 / Virus Database: 268.12.4/449 - Release Date: 15/09/2006
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/