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: variable transformation and centering


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: variable transformation and centering
Date   Mon, 4 Apr 2011 01:18:27 +0100

The marginal distribution of any predictor does not have implications
as such for regression.

Possibly, transforming skewed predictors may indirectly improve
matters in so far as assumptions of linearity, additivity and equal
variances are more nearly satisfied.

However, I suggest that your question answers itself if you focus on
one detail. From what you say, transformations such as logarithm or
square root are the most likely candidates. Translating your
predictors first by subtracting a summary will produce variables with
positive and negative values which will only be more difficult to
transform. So, don't centre first. Typically, centring will not be
necessary after transformation either.

On Sun, Apr 3, 2011 at 10:08 PM, Eduardo Nunez <enunezb@gmail.com> wrote:

> I have to run a regression model with several inflammatory biomarkers
> as dependent variables. They are continuous and heavily skewed
> variables.
> I would like to transform and center them before including in the model.
> My question is: should I transform the variable first and then center
> it? Or is preferred to center the variable first and then to transform
> it?
> Or any way doesn't make difference?
*
*   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