Calculate the number of permutations suitable for testing Multiscale
Codependence Analysis.
Usage
minpermute(alpha,nbtest,margin=1,ru=3)
Value
The minimum number of permutation to be used for
permute.cdp.
Arguments
alpha
The familywise type I error threshold allowable for the
complete analysis.
nbtest
The number of test performed (the number of
eigenvectors in the ‘mem’ object in the case of
MCA).
margin
A margin allowed for the number of permutation. Default
value: 1.
ru
The magnitude of the round-up to apply to the number of
permutations.
Author
Guillaume Guénard, Département des sciences biologiques,
Université de Montréal, Montréal, Québec, Canada.
Details
This function calculate the number of permutations for use with
permute.cdp. Parameter margin allows to apply a
safe margin to the number of permutations. The minimal suitable value
for this parameter is 1. Parameter ru allows one to round-up
the number of permutations. A value of 0 implies no round-up, a value
of 1 a round-up to the next ten, 2 a round-up to the next hundred, and
so on. Function minpermute is called internally by
permute.cdp in case permute = NA. In that case,
the margin is set to 10 (margin = 10) and the outcome is
rounded-up to the next thousand (ru = 3). This function is
meant for users that wish to apply their own margins and round-up
factors to calculate the number of permutations for use with
permute.cdp.
References
Guénard, G., Legendre, P., Boisclair, D., and Bilodeau, M. 2010.
Multiscale codependence analysis: an integrated approach to analyse
relationships across scales. Ecology 91: 2952-2964
# For a 5% threshold under 50 tests.minpermute(alpha = 0.05, nbtest=50)
# Allowing more margin (implies more computation time).minpermute(alpha = 0.05, nbtest=50, margin=10, ru=3)