Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

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


From   "Alvaro Zevallos" <azevallos@grade.org.pe>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: How can I make a graphic of regression results using only significant effects?
Date   Wed, 20 Sep 2006 09:52:28 -0500

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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index