[R-sig-eco] vegdist: "binomial", is it the scale-invariant version?
Pierre THIRIET
pierre.d.thiriet at gmail.com
Wed Mar 26 10:04:15 CET 2014
Dear all,
Please, could someone confirm that vegdist(X, meth="binomial") is the
binomial deviance *scaled* (i.e. the scale-invarient version of
"binomial deviance")
I want to use the dissimilarity "binomial deviance *scaled*", described
by Anderson and Millar (2004) as
where
"scaled" since it is the scale-invarient version of the "binomial
deviance" (by dividing by n/k/), also described by Anderson and Millar
(2004)
vegdist implements the "binomial index" as
d[jk] = sum*(*x[ij]*log(x[ij]/n[i]) + x[ik]*log(x[ik]/n[i]) -
n[i]*log(1/2)*)*/n[i] , where n[i] = x[ij] + x[ik]
It seems therefore that is the "binomial deviance scaled", whright ?
I thank you in advance,
cheers,
Pierre
Anderson, M.J. and Millar, R.B. (2004). Spatial variation and effects of
habitat on temperate reef fish assemblages in northeastern New Zealand.
/Journal of Experimental Marine Biology and Ecology/ 305, 191--221.
More information about the R-sig-ecology
mailing list