Re: [R] one/multi-dimensional scaling with incomplete dissimilarity matrix

From: Justin Donaldson <jjdonald_at_indiana.edu>
Date: Wed, 19 Mar 2008 14:12:16 -0400

Have you tried Gower dissimilarity or Sammon mapping? Gower dissimilarity is in the cluster package, under the 'daisy' method. Sammon mapping is in the MASS package, under the sammon method.

You may need to redo the dissimilarity matrix with Gower/daisy first, then Sammon. Keep in mind that the low dimensional representation is non-metric.

Your other option is to use 'imputed' distances, typically the mean values of the KNN of a missing value. This is often used in micro-array analyses of gene interactions. The package/method is 'impute'.

Of course, make sure that whatever method you use is appropriate for your data.

Best,
-Justin

On Wed, Mar 19, 2008 at 12:43 PM, Loic Thibaut <loic.thibaut_at_jcu.edu.au> wrote:

> Dear David,
>
> you asked this question a while ago on the R mailing list and got no
> answer. I have the same problem and was wondering if you had found a
> solution
>
> Cheers
>
> Loic
>
> Loic Thibaut, PhD candidate,
> ARC Centre of Excellence for Coral Reef Studies,
> School of Marine Biology, James Cook University,
> Townsville,
> Qld, 4811, Australia.
> Tel + 61 747 815 735, Fax: + 61 747 251 570, email:
> Loic.Thibaut_at_jcu.edu.au
>
> ______________________________________________
> R-help_at_r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

-- 
Justin Donaldson
PhD Candidate, Informatics
Indiana University
Department of Human Computer Interaction
Eigenmann Hall
1900 E.10th Street, Room 931
Bloomington, IN 47406

	[[alternative HTML version deleted]]

______________________________________________
R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Received on Wed 19 Mar 2008 - 18:16:40 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Wed 19 Mar 2008 - 19:31:15 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive