Statalist


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

Re: st: Basic regression interaction term question


From   "Frederick J. Boehmke" <frederick-boehmke@uiowa.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Basic regression interaction term question
Date   Sun, 12 Oct 2008 18:21:24 -0500 (CDT)

Others have given good advice on this question. There are multiple relevant questions that one can ask, including whether the coefficient on the interaction is different from zero or whether the marginal effect of B is different from zero. The latter includes two coefficients and depends on the value of A (H_0: b_B + b_AB*A = 0).

I'll use this opportunity to plug a utility I wrote to graph this marginal effect with a confidence interval. Type -net search grinter- to find it.

Description

grinter allows the user to graph the marginal effect of an interacted variable. grinter should work with any equation with a linear index: if the index is Xb = b1 + b2*x_1 + b3*x_2 + b4*x_1*x_2, grinter can graph either b2 + b4*x_2 or b3 + b4*x_1. The graph includes a vertical line at the mean value of the other constitutive term comprising the interaction (the modifying variable). It currently does not support other types of interactions (e.g., interactions with more than one variable).



On Fri, 10 Oct 2008, Michael I. Lichter wrote:

Date: Fri, 10 Oct 2008 13:53:11 -0400
From: Michael I. Lichter <mlichter@buffalo.edu>
Reply-To: statalist@hsphsun2.harvard.edu
To: statalist@hsphsun2.harvard.edu
Subject: st: Basic regression interaction term question

This is a pretty basic question, but I haven't been able to find any examples in the lit with this particular configuration ...

Suppose you regress Y on A and B, and you expect an interaction between A and B.

In the regression Y = A + B, the coefficient for B is not significant, but you have reason to think that it will be significant once you introduce the interaction term.

However, in the regression Y = A + B + AB, the coefficient for B remains non-significant even though the coefficient for AB is significant. Yet, "test A B AB" is significant.

Is it reasonable to treat this as a significant interaction?

What if AB is not significant either but "test A B AB" is still significant?

Michael
*
*   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/



                                                         - Fred

 < ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 >  Frederick J. Boehmke
 <  Associate Professor of Political Science
 >  University of Iowa
 <
 >  308 Schaeffer
 <  319-335-2342
 >  http://www.fredboehmke.net
   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
*
*   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   |   What's new   |   Site index