Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: What is the effect of centering on marginal effects?
From
Lisa Marie Yarnell <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: What is the effect of centering on marginal effects?
Date
Tue, 31 Jul 2012 19:53:01 -0700 (PDT)
Hi Stata users,
What is the effect of centering one's predictors on marginal effects? We centered our variables prior to cross-multiplication in the creation of interaction terms to avoid multicollinearity. We found that centering did not change the effect of the beta in our heckman model, but it eliminated the marginal effects produced by a mfx dyex command. Why would that be?
I don't think it's due to Stata estimating the marginal effects "at" another value because Stata estimates marginal effects at the mean by default, and the mean of the uncentered variable at hand (M = .29) is the equivalent of the mean of the centered variable (M = .00) because in centering the variable, we subtracted .29 from all scores.
Thanks,
Lisa
*
* 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/