Skip to content

center or scale before analyzing using pls.pcr

2 messages · Jinsong Zhao, Bjørn-Helge Mevik

#
Dear all,

I found pls.pcr package will give different results if the data are
centered and scaled using scale().

I am not sure about when I should scale my data, and whether the
dependent variable should be scaled. If the dependent variable is
scaled, how I give a prediction to the real data?

I appreciate for any suggestions and comments.

Best regards,

Jinsong


=====
(Mr.) Jinsong Zhao
Ph.D. Candidate
School of the Environment
Nanjing University
22 Hankou Road, Nanjing 210093
P.R. China
E-mail: jinsong_zh at yahoo.com
#
Jinsong Zhao <jinsong_zh at yahoo.com> writes:
Centering is done automatically by all implementations of PLSR I am
aware of (including pls.pcr, afaics).
There are no fixed rules about this.  Many practitioners live more or
less by the rule that unless the variables are `of the same type' or have
equal or comparable scales, they are scaled.  One example of data that
is typically not scaled (at least to begin with) is spectroscopic data.
There is no need for scaling the dependent variable.