[R] [R-pkgs] anacor: yet another ca package

Jan de Leeuw deleeuw at stat.ucla.edu
Sun Dec 23 18:57:08 CET 2007


anacor-0.9.0 is on CRAN (by De Leeuw and Mair)

anacor does correspondence analysis and canonical correspondence
analysis. It can make row plots, column plots,
joint plots, Benzécri plots, regression plots,
and transformation plots. Where appropriate, plots can
be in 3d using either rgl or scatterplot3d. Row and
column points can be in standard scaling, Benzécri
scaling, Goodman scaling, row-centroid, or column-centroid
scaling.

The summary method writes out a table with the chi-square
(inertia) decomposition, it also writes out the singular
values, and their asymptotic standard errors under multinomial
sampling. Plots of the category quantifications (row scores
and column scores) can be made optionally with asymptotic
confidence ellipses, again based on multinomial sampling.

The package contain various utilities to switch data formats,
in particular to transform data frames to Burt matrices,
to indicator matrices, and even to fuzzy indicator matrices
using B-spline bases.

The vignette for the package (not added yet) is a paper
also submitted to the special psychoR issue of JSS. You
can get a preprint at

http://idisk.mac.com/jdeleeuw-Public/psychoR/anacor.zip

The psychoR directory also has the anacor package, the
homals package, and the homals paper. smacof (many forms
of multidimensional scaling) is next.

==========================================================
Jan de Leeuw, 11667 Steinhoff Rd, Frazier Park, CA 93225
home 661-245-1725 mobile 661-231-5416 work 310-825-9550
.mac: jdeleeuw +++  aim: deleeuwjan +++ skype: j_deleeuw
==========================================================
          There is no worse screen to block out
    the spirit than confidence in our own intelligence.
                                    ---- John Calvin.

_______________________________________________
R-packages mailing list
R-packages at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-packages



More information about the R-help mailing list