Martin,
You got some fine response already from other r-users. Indeed,
centering explanatory variables is quite common and it really
can have benefits in numerical accuracy.
However one might also use centering for response variable and
even standardization i.e. x - mean(x) / sd(x). This
might increase the robustness of multivariate analysis, where
variables have very different scales. For example in biology -
agriculture: amount of milk of cows [ from few kg up to 50 kg or
even more ] and fat percentage in milk [ around 4% ]. Although,
centering of response variables is really not common.
--
Lep pozdrav / With regards,
Gregor GORJANC
---------------------------------------------------------------
University of Ljubljana
Biotechnical Faculty URI: http://www.bfro.uni-lj.si
Zootechnical Department email: gregor.gorjanc <at> bfro.uni-lj.si
Groblje 3 tel: +386 (0)1 72 17 861
SI-1230 Domzale fax: +386 (0)1 72 17 888
Slovenia
---------------------------------------------------------------
Message: 52
Date: Fri, 14 Jan 2005 16:44:21 -0500
From: "Martin Julien" <martin.julien.2 at courrier.uqam.ca>
Subject: [R] Centered variables and mixed-model
To: <r-help at stat.math.ethz.ch>
Message-ID: <200501142140.j0ELeJBQ028837 at intrant.uqam.ca>
Content-Type: text/plain; charset="iso-8859-1"
I work in biology and I use mixed-model for my data analysis
In a scientific paper, the author wrote:
"All continuous exploratory variables were centred on their median value
prior to inclusion in the analysis (Pinheiro & Bates, 2000)."
They refer to the book "Mixed-effects model in S and S-Plus" by
Pinheiro et
Bates in 2000.
I feel a bit strange with that paper because I can't find in the book why
they centred the variables on their median.
So I have two question:
First, is it correct to centred the variables on their median in a
mixed-model?
Second, why they do that?
Thank
Julien
