help-octave
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: QR vs LU factorisation


From: Dmitri A. Sergatskov
Subject: Re: QR vs LU factorisation
Date: Sun, 29 Jun 2008 14:12:45 -0500

On Sun, Jun 29, 2008 at 1:24 PM, Jordi Gutiérrez Hermoso
<address@hidden> wrote:
> This isn't specifically an Octave question except tangentially.
>
> I have some matrices that are as bad as can be: largish (1000x1000 or
> so), full, unsymmetric, and ill-conditioned. I notice that Octave uses
> LU factorisation with partial pivoting to invert these matrices as a
> last resort, which has been giving me acceptable results. Someone
> suggested to me that QR factorisation would be better suited. I'm
> reading Golub & Van Loan, but see no clear indication of when to use
> QR or LU.
>

I am not really an expert, but as far as I know

-- LU is a computationally efficient algorithm for "good" matrices, not
   suitable for ill-conditioned matrices at all.

-- SVD is a very robust algorithm; gives you the most "useful" results in
   the most pathological cases

-- QR seems to be good trade-off between the previous  two.
   As far as I know this is the default algorithm in both Matlab
   and IDl.

> Does anyone have any suggestions?
>
> Thanks,
> - Jordi G. H.
>

Regards,

Dmitri.
--



reply via email to

[Prev in Thread] Current Thread [Next in Thread]