[R] code for metric / nonmetric MDS

From: Jan de Leeuw <deleeuw_at_frazmtn.com>
Date: Sat 06 Aug 2005 - 09:44:42 EST

This does metric/nonmetric weighted least squares multidimensional scaling
using the smacof algorithm (deleeuw, 1977, in barra et al (eds), recent developments
in statistics; or deleeuw and heiser, 1980, in krishnaiah (ed), multivariate
analysis V). Uses a dist object for both dissimilarities and weights, which should both be non-negative for convergence. Fits euclidean distances only.

Monotone regression is done by pool-adjacent-violators, implemented in pava.R. The pava function can actually do monotone regression using weighted means, medians, or p-fractiles -- although for mds only means are used.

Somebody asked for this and it may be useful to others. It is not written
for efficiency, but the flexibility of allowing weights can be handy. By using relaxed smacof updates (see deleeuw and heiser or the book by borg and groenen) we can make this about twice as fast -- but that will come later. Feel free to tinker away.

Jan de Leeuw, 11667 Steinhoff Rd, Frazier Park, CA 93225, 661-245-1725
                                      Und die Einen sind im Dunkeln
                                      Und die Andern sind im Licht
                                      Und man siehe die im Lichte
                                      Die im Dunklen sieht man nicht


R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Sat Aug 06 09:49:04 2005

This archive was generated by hypermail 2.1.8 : Sun 23 Oct 2005 - 15:06:56 EST