cdt {compositions} | R Documentation |
Compute the centered default transform of a (data set of) compositions or amounts (or its inverse).
cdt(x,...) ## Default S3 method: cdt( x ,...) ## S3 method for class 'acomp': cdt( x ,...) ## S3 method for class 'rcomp': cdt( x ,...) ## S3 method for class 'aplus': cdt( x ,...) ## S3 method for class 'rplus': cdt( x ,...) ## S3 method for class 'rmult': cdt( x ,...) ## S3 method for class 'factor': cdt( x ,...) cdtInv(x,orig,...) ## Default S3 method: cdtInv( x ,orig,...) ## S3 method for class 'acomp': cdtInv( x ,orig,...) ## S3 method for class 'rcomp': cdtInv( x ,orig,...) ## S3 method for class 'aplus': cdtInv( x ,orig,...) ## S3 method for class 'rplus': cdtInv( x ,orig,...) ## S3 method for class 'rmult': cdtInv( x ,orig,...)
x |
a classed (matrix of) amount or composition, to be transformed with its centered default transform, or its inverse |
... |
generic arguments past to underlying functions. |
orig |
a compositional object which should be mimicked by the inverse transformation. It is used to determine the backtransform to be used and eventually to reconstruct the names of the parts. It is the generic argument. Typically this argument is the data set that has be transformed in the first place. |
The general idea of this package is to analyse the same data with
different geometric concepts, in a fashion as similar as possible. For each of the
four concepts there exists a unique transform expressing the geometry
in a linear subspace, keeping the relation to the variables. This
unique transformation is computed by cdt
. For
acomp
the transform is clr
, for
rcomp
it is cpt
, for
aplus
it is ilt
, and for
rplus
it is iit
. Each component of the result
is identified with a unit vector in the direction of the corresponding
component of the original composition or amount. Keep in mind that the
transform is not necessarily surjective and thus variances in the
image space might be singular.
A corresponding matrix or vector containing the transforms.
R. Tolosana-Delgado, K.Gerald v.d. Boogaart http://www.stat.boogaart.de
van den Boogaart, K.G. and R. Tolosana-Delgado (2008) "compositions": a unified R package to analyze Compositional Data, Computers & Geosciences, 34 (4), pages 320-338, doi:10.1016/j.cageo.2006.11.017.
## Not run: # the cdt is defined by cdt <- function(x) UseMethod("cdt",x) cdt.default <- function(x) x cdt.acomp <- clr cdt.rcomp <- cpt cdt.aplus <- ilt cdt.rplus <- iit ## End(Not run) x <- acomp(1:5) (ds <- cdt(x)) cdtInv(ds,x) (ds <- cdt(rcomp(1:5))) cdtInv(ds,rcomp(x)) data(Hydrochem) x = Hydrochem[,c("Na","K","Mg","Ca")] y = acomp(x) z = cdt(y) y2 = cdtInv(z,y) par(mfrow=c(2,2)) for(i in 1:4){plot(y[,i],y2[,i])}