[R-sig-eco] fourthcorner() with missing data

[Stéphane Dray]

I never used the function but from the doc, I deduced that 'recon' 
contains all reconstructed data while compleObs contains observed data 
and reconstructed data for missing observation. So, use the second one. 
Comparing both should help to evaluate the quality of the reconstruction.

Do not apologize. This question is very interersting and, to my 
knowledge, you would be the first to use fourthcorner whith missing values.

Cheers.


On 11/01/2013 13:50, Beth Atkinson wrote:
> Hi,
>
> Thanks for your help. I tried imputeAFDM() in missMDA, but when I 
> looked at completeObs the values of the original non-missing data were 
> changed for the quantitative trait. This doesn't seem to be correct.
>
> I'm new to trait analysis, so apologies if these are straightforward 
> issues.
>
> Beth
>
> --On 09 January 2013 17:04 +0100 Stéphane Dray 
> <stephane.dray at univ-lyon1.fr> wrote:
>
>> Hello,
>>
>> the method does not consider missing values. The best solution would be
>> to estimate missing values to fill the . For this estimation, you can
>> either use the nipals method (nipals function in ade4 with rec = 
>> TRUE) or
>> use the missMDA package. The second approach is better.
>>
>> Cheers.
>>
>> On 08/01/2013 16:40, Beth Atkinson wrote:
>>>
>>> Hello,
>>>
>>> I want to do a fourth corner analysis but I have missing data. I have
>>> 169 species from 32 plots. There are 4 habitat types and I'm looking
>>> at 8 traits. However, I don't have trait data for all species for each
>>> trait (each trait has data for at least 80% of species).
>>>
>>> The function fourthcorner in ade4 returns an error message about NAs.
>>> Is there a way around this? If I remove species that have data missing
>>> for any trait, I lose around a third of the species.
>>>
>>> Thanks,
>>>
>>> Beth
>>>
>>> --------------------------------------------------------
>>> Beth Atkinson
>>>
>>> PhD student
>>> Community Ecology Group
>>> School of Biological Sciences
>>> University of Bristol
>>> Woodland Road
>>> Bristol
>>> BS8 1UG
>>>
>>> http://www.bristol.ac.uk/biology/people/beth-atkinson/
>>>
>>> _______________________________________________
>>> R-sig-ecology mailing list
>>> R-sig-ecology at r-project.org
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>>>
>>
>>
>> -- 
>> Stéphane DRAY (stephane.dray at univ-lyon1.fr)
>> Laboratoire BBE-CNRS-UMR-5558, Univ. C. Bernard - Lyon I
>> 43, Bd du 11 Novembre 1918, 69622 Villeurbanne Cedex, France
>> Tel: 33 4 72 43 27 57       Fax: 33 4 72 43 13 88
>> http://pbil.univ-lyon1.fr/members/dray/
>>
>>
>>
>
>
>
> --------------------------------------------------------
> Beth Atkinson
>
> PhD student
> Community Ecology Group
> School of Biological Sciences
> University of Bristol
> Woodland Road
> Bristol
> BS8 1UG
>
> http://www.bristol.ac.uk/biology/people/beth-atkinson/
>
>
>
>
> .
>


-- 
Stéphane DRAY (stephane.dray at univ-lyon1.fr)
Laboratoire BBE-CNRS-UMR-5558, Univ. C. Bernard - Lyon I
43, Bd du 11 Novembre 1918, 69622 Villeurbanne Cedex, France
Tel: 33 4 72 43 27 57       Fax: 33 4 72 43 13 88
http://pbil.univ-lyon1.fr/members/dray/