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.
-------------- next part --------------
==Jan de Leeuw; Distinguished Professor and Chair, UCLA Department of
Statistics;
Editor: Journal of Multivariate Analysis, Journal of Statistical
Software
US mail: 8130 Math Sciences Bldg, Box 951554, Los Angeles, CA 90095-1554
phone (310)-825-9550; fax (310)-206-5658; email: deleeuw at stat.ucla.edu
.mac: jdeleeuw ++++++ aim: deleeuwjan ++++++ skype: j_deleeuw
homepages: http://gifi.stat.ucla.edu ++++++ http://www.cuddyvalley.org
------------------------------------------------------------------------
-------------------------
No matter where you go, there you are. --- Buckaroo Banzai
http://gifi.stat.ucla.edu/sounds/nomatter.au
------------------------------------------------------------------------
-------------------------