interpretation of interaction effect in regression
From: Ad de Jong (a.d.jong_at_tm.tue.nl)
Date: 02/23/05
- Next message: rbenton_at_ieee.org: "Re: Statistical Tests for Counts, varying size"
- Previous message: Mike Plichta: "autocorrelation as a predicor in multiple regression models"
- Messages sorted by: [ date ] [ thread ]
Date: Wed, 23 Feb 2005 13:36:29 +0000 (UTC)
Hi
Consider the following regression equations:
(1) Y= â0 + â1X + â2Z + â3XZ+ e.
(2) Y= â0 + â1X + â2X(quad) + â3Z + â4XZ+ e.
Equation 1 includes the linear terms of variables X and Z, and an
interaction term of X and Z. In Equation 2 the quadratic term
X(quad)is added. Does the inclusion of this quadratic term change the
interpretation of the interaction effect (XZ)?
I very appreciate your comments
Ad
- Next message: rbenton_at_ieee.org: "Re: Statistical Tests for Counts, varying size"
- Previous message: Mike Plichta: "autocorrelation as a predicor in multiple regression models"
- Messages sorted by: [ date ] [ thread ]
