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: What is the effect of centering on marginal effects?


From   Nick Winter <njgwinter@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: What is the effect of centering on marginal effects?
Date   Fri, 03 Aug 2012 09:47:57 -0400

Indeed.

I also wonder why people speak of "inflated" standard errors. The standard errors are correct when the data are (highly) correlated: they are telling you that the data don't contain much information on the independent effects of the correlated variables.

I've always enjoyed Goldberger's take on the "problem" of multicolinearity:

"Econometrics texts devote many pages to the problem of multicollinearity in multiple regression, but they say little about the closely analogous problem of small sample size in estimation a univariate mean. Perhaps that imbalance is attributable to the lack of an exotic polysyllabic name for 'small sample size'. If so, we can remove that impediment by introducing the term micronumerosity."

Goldberger, A. S. (1991). A Course in Econometrics. Harvard University Press, Cambridge MA.

Quoted at more length here: http://davegiles.blogspot.com/2011/09/micronumerosity.html


On 8/3/2012 8:13 AM, William Hauser wrote:
Dear all,

I'm fairly confident that mean centering does nothing to resolve
collinearity.  I believe it does fool some of the diagnostic tools
though and that's probably why the belief that it somehow solves the
problem persists.  Mean centering simply shifts the collinearity onto
the intercept term.  Mean centering adds no new information to the
model and that's the problem - the data lack the necessary information
for the model to partial out the effects in a precise and stable
manner.  Perhaps this is effect is different for interaction terms,
but I fail to see how that's the case.

Collinearity means the independent effects of the collinear variables
cannot be precisely estimated.  The point of interaction terms is that
they be analyzed jointly anyway.  The use of the margins and
marginsplot commands accomplish this with such ease (and polish) that
I would heartily recommend their use.

Will
*
*   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/


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