Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Test for significance in Interaction model


From   David Hoaglin <dchoaglin@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Test for significance in Interaction model
Date   Sun, 3 Jun 2012 13:48:13 -0400

Dear Lukas,

It would help if you told us more about what you mean by "the combined
effect of x2 and x1*x2 for the case of x1=1."

In your simplified model,

b0 is the mean of y for x1 = 0, adjusting for the contribution of x2.
In this model, when x1 = 0, you have the line y = b0 + b2x2, so b0 is
its intercept.

b1 is the difference between the mean of y for x1 = 1 and the mean of
y for x1 = 0, adjusting for the contribution of x2.

b2 is the slope of y against x2 when x1 = 0

b2 + b3 is the slope of y against x2 when x1 = 1.


It is often of interest to test whether b3 = 0 (i.e., whether the
slope of y against x2 is the same for x1 = 1 and x1 = 0) and, if so,
to test whether b2 = 0 and b3 = 0 (i.e., no slope against x2).

David Hoaglin

On Sun, Jun 3, 2012 at 7:04 AM, Lukas Borkowski <lukasbork@hotmail.com> wrote:
> Dear all,
>
> I am a quite new stata user and I am trying to find a way to check whether the combined effect of my interaction effect is significant.
> My model can be simplified as y = b0 + b1x1 + b2x2 + b3x*1x2.  x1 is a dummy variable and x2 a  variable. I would now like to check whether the combined effect of x2 and x1*x2 for the case of x1=1 is significant. But I am not sure how. I used -test x2 x1*x2- and -testnl _b[x2]*_b[x1*x2] = 0- but I have absolutely no idea if that is correct or not. Could you please help me? Thank you!

*
*   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/


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