References & Citations
Condensed Matter > Other Condensed Matter
Title: Efficient Rank Reduction of Correlation Matrices
(Submitted on 18 Mar 2004 (v1), last revised 25 Jan 2005 (this version, v2))
Abstract: Geometric optimisation algorithms are developed that efficiently find the nearest low-rank correlation matrix. We show, in numerical tests, that our methods compare favourably to the existing methods in the literature. The connection with the Lagrange multiplier method is established, along with an identification of whether a local minimum is a global minimum. An additional benefit of the geometric approach is that any weighted norm can be applied. The problem of finding the nearest low-rank correlation matrix occurs as part of the calibration of multi-factor interest rate market models to correlation.
Submission history
From: Raoul Pietersz [view email][v1] Thu, 18 Mar 2004 19:33:25 GMT (233kb)
[v2] Tue, 25 Jan 2005 11:48:07 GMT (73kb)
Link back to: arXiv, form interface, contact.