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 on April 23, and its replacement, is already up and running.

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

Re: st: Quadratic Interaction Term.

From   Maarten buis <>
Subject   Re: st: Quadratic Interaction Term.
Date   Mon, 13 Sep 2010 08:16:07 +0000 (GMT)

--- On Fri, 10/9/10, natasha agarwal wrote:
> I want to interact FDI*GDP and then introduce the quadratic
> term of GDP with its interaction with FDI. So now the estimating
> equation looks like
> lnVA=lnK+Lnl+lnFDI+lnGDP+lnGDP^2+lnFDI*lnGDP+lnFDI*(lnGDP^2)
> When I estimate this equation, I get the coefficient on
> the lnFDI*(lnGDP^2) is insignificant but when I remove lnGDP^2
> and re-estimate the same lnFDI*(lnGDP^2) becomes significant.
> However, the sign and magnitude does not change. The sign,
> Can anyone please explain me why is this happening?

Now the interaction term is picking up non-linearity that actually
occured in the main effect. In general, if you want to add 
non-linearity in combination with an interaction effect, then I
would make sure that the main effect is at least as flexible as
the interaction effect to avoid this problem.

Hope this helps,

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen


*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index