Performs classical or nonmetric Multidimensional Scaling analysis of provenance data
MDS(x, ...)# S3 method for default
MDS(x, classical = FALSE, k = 2, ...)
# S3 method for compositional
MDS(x, classical = FALSE, k = 2, ...)
# S3 method for counts
MDS(x, classical = FALSE, k = 2, ...)
# S3 method for distributional
MDS(x, classical = FALSE, k = 2, nb = 0, ...)
# S3 method for varietal
MDS(x, classical = FALSE, k = 2, nb = 0, ...)
an object of class MDS, i.e. a list containing the
following items:
points: a two column vector of the fitted configuration
classical: a boolean flag indicating whether the MDS
configuration was obtained by classical (TRUE) or nonmetric
(FALSE) MDS.
diss: the dissimilarity matrix used for the MDS analysis
stress: (only if classical=TRUE) the final stress
achieved (in percent)
an object of class distributional,
compositional, counts, varietal or
diss
optional arguments
If x has class distributional, ... is passed
on to diss.distributional.
If x has class compositional, ... is passed on
to diss.compositional.
If x has class counts, ... is passed on to
diss.counts.
If x has class varietal, ... is passed on to
diss.varietal.
Otherwise, ... is passed on to cmdscale (if
classical=TRUE), to isoMDS (if
classical=FALSE).
boolean flag indicating whether classical
(TRUE) or nonmetric (FALSE) MDS should be used
the desired dimensionality of the solution
number of bootstrap resamples. If nb>0, then
plot.MDS(...) will visualise the sampling uncertainty as
polygons (inspired by Nordsvan et al. 2020). The bigger
nb, the slower the calculations. nb=10 seems a
good compromise.
Nordsvan, A.R., Kirscher, U., Kirkland, C.L., Barham, M. and Brennan, D.T., 2020. Resampling (detrital) zircon age distributions for accurate multidimensional scaling solutions. Earth-Science Reviews, p.103149.
Vermeesch, P., 2013, Multi-sample comparison of detrital age distributions. Chemical Geology v.341, 140-146, doi:10.1016/j.chemgeo.2013.01.010
data(Namib)
plot(MDS(Namib$Major,classical=TRUE))
Run the code above in your browser using DataLab