Skip to content
Prev 206557 / 398503 Next

optimization problem

Klaus,

I am happy to know that the quadratic cost LSAP seems to work well for you.  

The Hungarian algorithm is a classic for solving linear sum assignment problem, which is closely related to matching in bipartite graphs.  You can google or wiki these terms to get papers and books on this topic. 

Also look at the PhD thesis by one of Nick Higham's students (Section 3.5):

 Parallel Solution of SVD-Related Problems, With Applications

at

http://www.maths.manchester.ac.uk/~higham/misc/past-students.php

This thesis has a good discussion of different variants of your problem, some of which are more general than yours.  

Your problem is one of the variants of "procrustes-type" problems found in multivariate statistics.  The very same problem that you posed is supposed to occur in multidimensional scaling.  So, you might also want to look in that literature.


Best,
Ravi.

____________________________________________________________________

Ravi Varadhan, Ph.D.
Assistant Professor,
Division of Geriatric Medicine and Gerontology
School of Medicine
Johns Hopkins University

Ph. (410) 502-2619
email: rvaradhan at jhmi.edu


----- Original Message -----
From: klausch at gmx.de
Date: Sunday, January 17, 2010 8:06 am
Subject: Re: [R] optimization problem
To: Ravi Varadhan <rvaradhan at jhmi.edu>, erwin.kalvelagen at gmail.com, hwborchers at googlemail.com
Cc: r-help at stat.math.ethz.ch