[R-sig-eco] Re : Problem to use varpart

Eduard Szöcs szoe8822 at uni-landau.de
Thu Jul 7 15:36:40 CEST 2011


Hai Momadou,

have a look at the relaimpo-package.


HTH,

Edi


Am 07.07.2011 14:32, schrieb Gavin Simpson:
> On Thu, 2011-07-07 at 12:26 +0100, momadou sow wrote:
>> Hi,
>>
>> Thank Gavin for your answer but for your request “but why would you
>> want to?” : to partition the explained variance in my linear
>> regression. Thus, I want have the contribution of each explanatory
>> variable.
>> Momadou
> Don't cherry-pick. What I actually said was why would you want to use
> RDA to fit a linear model? Not why do you want to do variance
> partitioning.
>
> Anyway, as I said, you can't do what you want with varpart() so you need
> to look at how you decompose a linear model (lm()).
>
> G
>> De : Gavin Simpson<gavin.simpson at ucl.ac.uk>
>> À : momadou sow<momadou at yahoo.fr>
>> Cc : "R-sig-ecology at r-project.org"<R-sig-ecology at r-project.org>
>> Envoyé le : Jeudi 7 Juillet 2011 23h00
>> Objet : Re: [R-sig-eco] Problem to use varpart
>>
>> On Thu, 2011-07-07 at 06:33 +0100, momadou sow wrote:
>>> Hi,
>>> I did a linear regression with 5 explanatory variables. Now, to see
>> the contribution of each variable, I use varpart  from vegan library.
>> But varpart don’t accepts my 5 explanatory variables, it accept only
>> 4.
>>> 1- How must I  do to use my 5 explanatory variables?
>>> 2- Is it the sum of variance fraction of each variable must be equal
>> to 1.
>>> Thanks for your help.
>> varpart is for redundancy analysis - aka reduced rank regression. I
>> suppose one can do linear regression with RDA, but why would you want
>> to?
>>
>> Anyway, you can't do what you want with varpart(). Look at
>> interpretting
>> the linear model directly.
>>
>> G
>>
>> -- 
>> %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
>> Dr. Gavin Simpson            [t] +44 (0)20 7679 0522
>> ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
>> Pearson Building,            [e] gavin.simpsonATNOSPAMucl.ac.uk
>> Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
>> UK. WC1E 6BT.                [w] http://www.freshwaters.org.uk
>> %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
>>
>>
>>
>>



More information about the R-sig-ecology mailing list