Path: news.mathworks.com!not-for-mail
From: <HIDDEN>
Newsgroups: comp.soft-sys.matlab
Subject: Re: Matrix inversion
Date: Sat, 1 Nov 2008 02:56:03 +0000 (UTC)
Organization: The MathWorks, Inc.
Lines: 9
Message-ID: <geggg3$p3s$1@fred.mathworks.com>
References: <geft5a$bbp$1@fred.mathworks.com>
Reply-To: <HIDDEN>
NNTP-Posting-Host: webapp-02-blr.mathworks.com
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-Trace: fred.mathworks.com 1225508163 25724 172.30.248.37 (1 Nov 2008 02:56:03 GMT)
X-Complaints-To: news@mathworks.com
NNTP-Posting-Date: Sat, 1 Nov 2008 02:56:03 +0000 (UTC)
X-Newsreader: MATLAB Central Newsreader 1187260
Xref: news.mathworks.com comp.soft-sys.matlab:498377


"Katula " <ssengendo_ronald@yahoo.co.uk> wrote in message <geft5a$bbp$1@fred.mathworks.com>...
> Am trying to learn MATLAB,i formed a matrix which has large values,zeros and ones,when i try to get the inverse it says matrix is singular.Question is how can i combine very small values with very large values to form one matrix?

  There is no reason why a non-singular matrix cannot have both very small and very large values at the same time.  It all depends on what those values are.

  When Matlab declares that it deems a matrix to be "singular" it means that within computable accuracy the matrix has a zero value for its determinant, which in turn means that it can possess no inverse.  If it does in fact have an inverse in a mathematical sense, Matlab is unable to determine that inverse with any reasonable accuracy because of inherent round-off errors.

Roger Stafford