[R] simpleboot, pairs.boot

[Rebecca Young]

Hi.

I am a new R user trying to obtain bootstrap confidence intervals around vector
correlations of Principal components.

My data is a data frame of eigenvector loadings from PC1 in the columns the rows
are different species.  I want to calculated the vector correlation and
bootstrap confidence intervals for all species pairs.

I have tried using simpleboot as pairs.boot seems to be an easy way of sampling
from two vectors of data.  The function I wrote "veccor" below works outside of
the pairs.boot function.  However in this function it seems to calculate all
1000 vector correlations as -1.

########
veccor
function(x,y) ((sum((x)*(y)))/((sqrt(sum(y^2)))*(sqrt(sum(x^2)))))
x<-as.vector(faVC[1,])
y<-as.vector(faVC[2,])
boot<-pairs.boot(x,y,veccor,1000)
#########

I have also tried using the boot function but only end up sampling from on of my
vectors of data.

Any suggestions would be greatly appriciated!  Thanks!

Best, Becca