The Open UniversitySkip to content
 

The ℓ 1 oblique procrustes problem

Trendafilov, Nickolay T. and Watson, G. A. (2004). The ℓ 1 oblique procrustes problem. Statistics and Computing, 14(1) pp. 39–51.

Full text available as:
Full text not publicly available
Due to copyright restrictions, this file is not available for public download
DOI (Digital Object Identifier) Link: http://dx.doi.org/10.1023/B:STCO.0000009415.14785.2a
Google Scholar: Look up in Google Scholar

Abstract

In this paper, we reconsider the well-known oblique Procrustes problem where the usual least-squares objective function is replaced by a more robust discrepancy measure, based on the ℓ1 norm or smooth approximations of it. We propose two approaches to the solution of this problem. One approach is based on convex analysis and uses the structure of the problem to permit a solution to the ℓ1 norm problem. An alternative approach is to smooth the problem by working with smooth approximations to the ℓ 1 norm, and this leads to a solution process based on the solution of ordinary differential equations on manifolds. The general weighted Procrustes problem (both orthogonal and oblique) can also be solved by the latter approach. Numerical examples to illustrate the algorithms which have been developed are reported and analyzed.

Item Type: Journal Article
Copyright Holders: 2004 Kluwer Academic Publishers
ISSN: 1573-1375
Keywords: fitting configurations; constrained optimization; dynamical system on manifolds; descent flows; optimality conditions; reference-structure and factor-pattern
Academic Unit/Department: Mathematics, Computing and Technology > Mathematics and Statistics
Item ID: 25481
Depositing User: Nickolay Trendafilov
Date Deposited: 31 Mar 2011 14:51
Last Modified: 25 May 2011 01:07
URI: http://oro.open.ac.uk/id/eprint/25481
Share this page:

Actions (login may be required)

View Item
Report issue / request change

Policies | Disclaimer

© The Open University   + 44 (0)870 333 4340   general-enquiries@open.ac.uk