R help - Mar 2004 - boot package

Dear all 

As part of an ongoing study on the ecomorphology of ant communities, I have 
obtained a matrix with 156 row (species) and 20 columns (several measurements 
of body shape) for 4 localities.

For each community, I calculated a matrix of Euclidean distances between all
pairs of species. From this matrix, I extracted two measures of community
structure: i) I identified the distance from a individual to its nearest
neighbor (NND) in the morphological space and then calculated the averages of 
the NND (MNND); ii) the mean of the Euclidean distances (MED). NND and MED 
are of practical use in describing spatial relations between species. The 
results of Euclidean distance studies in each of the four localities were 
compared with each other for evidence repeating patterns.

To determine wheter ou not the morphological arrangement of species in 
communities reflected internal organization, MNND and MED should be compared 
with randomly generated communities.

It is possible to write a function in th boot package to evaluate the values 
of average nearest-neighbor distance and of the mean of Euclidean distances 
in randomly generated communities ? One set of 1000 random communities each 
with 78, 90, 100 and 102 species must be generaded from the pool of species 
(156 species). The only restrictions on species composition  is that no 
species could be placed in a community more than once, and only rows values 
can be permuted.

Sorry, I don't know how to define the statistic term in the boot() function 
from the boot library in R. I have studied the boot package, argument list of 
functions and the functions definition. Can someone point me the correct 
syntax?

I would very much appreciate any help

Best regards

Rog?rio

Hi,
The function anosim() in vegan package or sample() in base may be of help to
you.
Alex

-----Original Message-----
From: Rog?rio Rosa da Silva [mailto:rrsilva at ib.usp.br] 
Sent: March 4, 2004 4:57 AM
To: rhelp
Subject: [R] boot package 



Dear all 

As part of an ongoing study on the ecomorphology of ant communities, I have 
obtained a matrix with 156 row (species) and 20 columns (several
measurements 
of body shape) for 4 localities.

For each community, I calculated a matrix of Euclidean distances between all
pairs of species. From this matrix, I extracted two measures of community
structure: i) I identified the distance from a individual to its nearest
neighbor (NND) in the morphological space and then calculated the averages
of 
the NND (MNND); ii) the mean of the Euclidean distances (MED). NND and MED 
are of practical use in describing spatial relations between species. The 
results of Euclidean distance studies in each of the four localities were 
compared with each other for evidence repeating patterns.

To determine wheter ou not the morphological arrangement of species in 
communities reflected internal organization, MNND and MED should be compared

with randomly generated communities.

It is possible to write a function in th boot package to evaluate the values

of average nearest-neighbor distance and of the mean of Euclidean distances 
in randomly generated communities ? One set of 1000 random communities each 
with 78, 90, 100 and 102 species must be generaded from the pool of species 
(156 species). The only restrictions on species composition  is that no 
species could be placed in a community more than once, and only rows values 
can be permuted.

Sorry, I don't know how to define the statistic term in the boot() function 
from the boot library in R. I have studied the boot package, argument list
of 
functions and the functions definition. Can someone point me the correct 
syntax?

I would very much appreciate any help

Best regards

Rog?rio

______________________________________________
R-help at stat.math.ethz.ch mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html