@ Stas
Thanks for the extensive answer! I squeezed my data in your function but
still need to mull over it and your comments for some time.
спасибо,
Axel
Stas Kolenikov schrieb:
I don't know if there are bugs in the code, but the step 4) does not
compute significance... at least the way s
I don't know if there are bugs in the code, but the step 4) does not
compute significance... at least the way statisticians know it. The
fractions above or below 0 are not significance. I don't even know how
to call those... probably cdf of the bootstrap distribution evaluated
at zero.
Let's put t
Hi Alex,
I presume you've asked Jari what the bug is? He obviously knows and you
are asking a lot for the rest of us to i) know the method you speak of
and ii) debug the code of another.
As an alternative, try function testdim() in the ade4 package. This
implements the method of Dray:
Dray, S (2
Axel Strauß schrieb:
G'Day R users!
Following an ordination using prcomp,
Sorry, correction. I mean "using princomp".
I'd like to test which variables singnificantly contribute to a
principal component. There is a method suggested by Peres-Neto and al.
2003. Ecology 84:2347-2363 called "b
G'Day R users!
Following an ordination using prcomp, I'd like to test which variables
singnificantly contribute to a principal component. There is a method
suggested by Peres-Neto and al. 2003. Ecology 84:2347-2363 called
"bootstrapped eigenvector". It was asked for that in this forum in
Jan
5 matches
Mail list logo
