Statalist


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

Re: AW: AW: st: AW: beta coefficients for interaction terms


From   Ulrich Kohler <kohler@wzb.eu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: AW: AW: st: AW: beta coefficients for interaction terms
Date   Mon, 27 Apr 2009 11:13:14 +0200

I wasn't in my office the last couple of days so I could not commend on
this. As originator of the confusion let me add that the term -ia2- in
Kit Baum's solution is formed by the standardized (and therefore
centered) constituent variables -shead- and -slength-. You would not
have gotten the same solution if you had used the raw values of the
constituent variables for forming the interaction term, i.e.

. sysuse auto, clear
. gen ia1 = mpg*length
. reg mpg headroom length ia1, beta

is not equal to 

. egen shead = std(headroom)
. egen slength = std(length)
. gen ia2 = shead*slength
. reg mpg headroom length ia2, beta

But the latter is equal to 

. egen smpg = std(mpg)
. egen sia2 = std(ia2)
. reg smpg sheadroom slength sia2, beta

(In this last command you will see that the estimated coefficients and
the standardized coefficients are the same)


Uli

Am Dienstag, den 21.04.2009, 18:10 +0200 schrieb
lschoele@rumms.uni-mannheim.de:
> I got different results first, but trying it again gave me the same results.
> According to the book "Data analysis with stata" from Kohler and  
> Kreuter, you are not supposed to use
> reg mpg headroom length ia2, beta
> you are supposed to use
> reg smpg shead slength sia2
> for interaction effects, but if I get the same beta coefficients, why  
> is it wrong to use reg mpg headroom length ia2, beta? That confuses me  
> a little.
> 
> 
> Zitat von Martin Weiss <martin.weiss1@gmx.de>:
> 
> > <>
> >
> >
> > How do you mean that? The example that Kit provided gives the same results?
> > Did you get something else for your special dataset?
> >
> >
> >
> > HTH
> > Martin
> >
> >
> > -----Ursprüngliche Nachricht-----
> > Von: owner-statalist@hsphsun2.harvard.edu
> > [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von
> > lschoele@rumms.uni-mannheim.de
> > Gesendet: Dienstag, 21. April 2009 16:10
> > An: statalist@hsphsun2.harvard.edu
> > Betreff: re: AW: st: AW: beta coefficients for interaction terms
> >
> > I tried it out and got different figures for the beta coefficient.
> > Thats why I asked that question.
> > It should give me the same figures, right?
> >
> > Zitat von Christopher Baum <baum@bc.edu>:
> >
> >> <>
> >> Why don't you just try it out?
> >>
> >> sysuse auto, clear
> >> egen shead = std(headroom)
> >> egen slength = std(length)
> >> egen smpg = std(mpg)
> >> gen ia2 = shead*slength
> >> egen sia2 = std(ia2)
> >> reg smpg shead slength sia2
> >> reg mpg headroom length ia2, beta
> >>
> >> Kit Baum   |   Boston College Economics and DIW Berlin   |
> >> http://ideas.repec.org/e/pba1.html
> >> An Introduction to Stata Programming   |
> >> http://www.stata-press.com/books/isp.html
> >> An Introduction to Modern Econometrics Using Stata   |
> >> http://www.stata-press.com/books/imeus.html
> >>
> >> *
> >> *   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/
> >
> >
> > *
> > *   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/
> >
> >
> > *
> > *   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/
> >
> >
> 
> 
> 
> *
> *   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/
-- 
kohler@wzb.eu
030 25491-361


*
*   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