complex SVD

[Gary Ballantyne]

Hi All

Please correct me if I'm wrong, but it seems like a SVD of a complex matrix is not available (i.e., no T.svd).

If so, is the best approach to use an "equivalent real formulation" ( http://www.osti.gov/scitech/biblio/756121 ) ?

E.g., say C = A + jB is complex, we instead deal with [A -B;B A] (in matlab notation).

Or perhaps some other cunning approach?

Cheers

Gary

[Sébastien Loisel]

Hi,

You are right, I didn't implement complex SVD. You might be able to find an equivalent real formulation!

--
You received this message because you are subscribed to the Google Groups "numericjs" group.
To unsubscribe from this group and stop receiving emails from it, send an email to numericjs+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
Sébastien Loisel

Department of Mathematics
Heriot-Watt University
Riccarton EH14 4AS
United Kingdom
web: http://www.ma.hw.ac.uk/~loisel/
email: S.Loisel at hw.ac.uk
phone: +44 131 451 3234
fax: +44 131 451 3249

[Gary Ballantyne]

Thanks Sébastien

I have a complex SVD working with an equivalent real matrix and your SVD
(and coffeescript) here: http://puzlet.com/m/b00cf .

On 22/03/14 00:53, Sébastien Loisel wrote:
> Hi,
>
> You are right, I didn't implement complex SVD. You might be able to find
> an equivalent real formulation!
>
>
> On Thu, Mar 20, 2014 at 7:33 PM, Gary Ballantyne
> <gary.ba...@gmail.com <mailto:gary.ba...@gmail.com>> wrote:
>
> Hi All
>
> Please correct me if I'm wrong, but it seems like a SVD of a complex
> matrix is not available (i.e., no T.svd).
>
> If so, is the best approach to use an "equivalent real formulation"
> ( http://www.osti.gov/scitech/biblio/756121 ) ?
>
> E.g., say C = A + jB is complex, we instead deal with [A -B;B A] (in
> matlab notation).
>
> Or perhaps some other cunning approach?
>
> Cheers
>
> Gary
>
> --
> You received this message because you are subscribed to the Google
> Groups "numericjs" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to numericjs+...@googlegroups.com

> <mailto:numericjs+...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout.
>
>
>
>
> --
> Sébastien Loisel
> Department of Mathematics
> Heriot-Watt University
> Riccarton EH14 4AS
> United Kingdom
> web: http://www.ma.hw.ac.uk/~loisel/

> email: S.Loisel at hw.ac.uk <http://hw.ac.uk>

> phone: +44 131 451 3234
> fax: +44 131 451 3249
>

> --
> You received this message because you are subscribed to the Google
> Groups "numericjs" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to numericjs+...@googlegroups.com

> <mailto:numericjs+...@googlegroups.com>.

[Sébastien Loisel]

Thanks!

On Sun, Mar 23, 2014 at 10:17 PM, Gary Ballantyne <gary.ba...@gmail.com> wrote:

Thanks Sébastien

I have a complex SVD working with an equivalent real matrix and your SVD (and coffeescript) here: http://puzlet.com/m/b00cf .

On 22/03/14 00:53, Sébastien Loisel wrote:

Hi,

You are right, I didn't implement complex SVD. You might be able to find
an equivalent real formulation!


On Thu, Mar 20, 2014 at 7:33 PM, Gary Ballantyne

<gary.ba...@gmail.com <mailto:gary.ballantyne@gmail.com>> wrote:

    Hi All

    Please correct me if I'm wrong, but it seems like a SVD of a complex
    matrix is not available (i.e., no T.svd).

    If so, is the best approach to use an "equivalent real formulation"
    ( http://www.osti.gov/scitech/biblio/756121 ) ?

    E.g., say C = A + jB is complex, we instead deal with [A -B;B A] (in
    matlab notation).

    Or perhaps some other cunning approach?

    Cheers

    Gary

    --
    You received this message because you are subscribed to the Google
    Groups "numericjs" group.
    To unsubscribe from this group and stop receiving emails from it,

    send an email to numericjs+unsubscribe@googlegroups.com

    <mailto:numericjs+unsubscribe@googlegroups.com>.

    For more options, visit https://groups.google.com/d/optout.




--
Sébastien Loisel
Department of Mathematics
Heriot-Watt University
Riccarton EH14 4AS
United Kingdom
web: http://www.ma.hw.ac.uk/~loisel/

email: S.Loisel at hw.ac.uk <http://hw.ac.uk>

phone: +44 131 451 3234
fax: +44 131 451 3249

--
You received this message because you are subscribed to the Google
Groups "numericjs" group.
To unsubscribe from this group and stop receiving emails from it, send

an email to numericjs+unsubscribe@googlegroups.com

<mailto:numericjs+unsubscribe@googlegroups.com>.

For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to the Google Groups "numericjs" group.

To unsubscribe from this group and stop receiving emails from it, send an email to numericjs+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
Sébastien Loisel

Department of Mathematics
Heriot-Watt University
Riccarton EH14 4AS
United Kingdom
web: http://www.ma.hw.ac.uk/~loisel/
email: S.Loisel at hw.ac.uk

[Johnson]

Thank you!

I have read your post and tried to implement accordingly. Nevertheless, I found it difficult when identical singular values are present, as I am not sure how to extract the real and imaginary entries since the singlular value cannot be distinguished. Any suggestions?

Thank you!