R help - Sep 2011 - calc.relimp pmvd for US R-user

Dear All:

I am calculating  the relative importance of a regressor in a linear model. 
Does anyone know how I can obtain/install the 'pmvd' computation type? I
am
a US user.

Regards,
Y 

--
View this message in context:
http://r.789695.n4.nabble.com/calc-relimp-pmvd-for-US-R-user-tp3808752p3808752.html
Sent from the R help mailing list archive at Nabble.com.

YAddo <linkyox <at> gmail.com> writes:
> 
> Dear All:
> 
> I am calculating  the relative importance of a regressor in a linear model.
> Does anyone know how I can obtain/install the 'pmvd' computation
type? I am
> a US user.
> 
  I didn't know what the heck you were talking about, but having looked
at http://prof.beuth-hochschule.de/groemping/relaimpo/ it seems that
you can't legally install that version of the package because of
conflicts with <http://www.patentstorm.us/patents/6640204.html>.

  Would some of the other results of 
library("sos")
findFn("{relative importance}")

work for you?

  Ben Bolker

> Dear All:
>
> I am calculating  the relative importance of a regressor in a linear
> model.
> Does anyone know how I can obtain/install the 'pmvd' computation
type? I
> am
> a US user.
>
> Regards,
> Y
>
Hi,

Do you have any specific reason to use the "pmvd" method, especially
in
the light of
http://prof.beuth-hochschule.de/fileadmin/user/groemping/downloads/amstat07mayp139.pdf?
The default "lmg" method
- gives very similar results,
- is more well-known in the literature, and
- is included in the US-version of the package.
Anyway, as a non-US user once upon a time I tried both methods, and they
really did not deviate that much (at least the ordering of the variables
was highly similar).


Hope that helps,
   Denes




> --
> View this message in context:
>
http://r.789695.n4.nabble.com/calc-relimp-pmvd-for-US-R-user-tp3808752p3808752.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>