| Determinant of a matrix [message #86569] |
Wed, 20 November 2013 04:09  |
fd_luni
Messages: 66
Registered: January 2013
|
Member |
|
|
Hi
I want to check the linear independence of the columns on my matrix. I found the function DETERM() but this works only in cases where the matrix is NXN. In my case the matrix is NX4 (N-rows, 4-columns).
How can I calculate the determinant of that matrix?
Many Thanks
Ma
|
|
|
|
|
|
|
|
| Re: Determinant of a matrix [message #86574 is a reply to message #86569] |
Wed, 20 November 2013 05:37  |
lecacheux.alain
Messages: 325
Registered: January 2008
|
Senior Member |
|
|
Le mercredi 20 novembre 2013 13:09:48 UTC+1, fd_...@mail.com a écrit :
> Hi
>
>
>
> I want to check the linear independence of the columns on my matrix. I found the function DETERM() but this works only in cases where the matrix is NXN. In my case the matrix is NX4 (N-rows, 4-columns).
>
>
>
> How can I calculate the determinant of that matrix?
>
>
>
> Many Thanks
>
> Ma
> How can I calculate the determinant of that matrix?
You cannot, because determinant is not defined for a non square matrix.
> I want to check the linear independence of the columns on my matrix
What you are asking for is the rank of your matrix (i.e., mathematically, the number of independent columns). Matrix rank can be determined by singular value decomposition: the rank is the number of singular values which are not zero. In IDL, you can write:
if A is your matrix:
IDL> LA_SVD, A, W, U, V
IDL> rank_of_A = N_elements(W[where(W ne 0)])
alx.
|
|
|
|
|
|
|
|
|
|
|
|
| Re: Determinant of a matrix [message #86579 is a reply to message #86578] |
Wed, 20 November 2013 08:33 |
Craig Markwardt
Messages: 1869
Registered: November 1996
|
Senior Member |
|
|
On Wednesday, November 20, 2013 7:53:22 AM UTC-8, fd_...@mail.com wrote:
>> Maria, are you going to delete your posts again?
>
>
>
> I had delete my post before because instead of singular values I wrote eigenvalues which is wrong designation. I didn't know how to edit my post that is why I had delete it and re post it correctly. I didn't know that we are not allowed to delete our posts. Sorry about that.
Well, you deleted two posts in your "Derivatives" thread, not just one.
You can do whatever you want, but if you delete your posts, I'm going to recommend that people don't respond. Deleting one's posts is unfair to other readers and to people that try to respond to the post.
Nobody can edit posts. If you make an error - that's fine, everybody does - then just post a follow-up correcting the error.
CM
|
|
|
|
| Re: Determinant of a matrix [message #86580 is a reply to message #86579] |
Wed, 20 November 2013 08:46 |
Mats Löfdahl
Messages: 263
Registered: January 2012
|
Senior Member |
|
|
Den onsdagen den 20:e november 2013 kl. 17:33:20 UTC+1 skrev Craig Markwardt:
> On Wednesday, November 20, 2013 7:53:22 AM UTC-8, fd_...@mail.com wrote:
>
>>> Maria, are you going to delete your posts again?
>
>> I had delete my post before because instead of singular values I wrote eigenvalues which is wrong designation. I didn't know how to edit my post that is why I had delete it and re post it correctly. I didn't know that we are not allowed to delete our posts. Sorry about that.
>
> Well, you deleted two posts in your "Derivatives" thread, not just one.
>
> You can do whatever you want, but if you delete your posts, I'm going to recommend that people don't respond. Deleting one's posts is unfair to other readers and to people that try to respond to the post.
>
> Nobody can edit posts. If you make an error - that's fine, everybody does - then just post a follow-up correcting the error.
Deleting posts in google groups is also not very effective. They may disappear from google but many proper usenet news servers ignore such deletes. Which is another reason to do corrections in followups rather than by trying to replace posts with new versions.
|
|
|
|
| Re: Determinant of a matrix [message #86582 is a reply to message #86580] |
Wed, 20 November 2013 09:04 |
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
On 11/20/13 11:46, Mats Löfdahl wrote:
> Den onsdagen den 20:e november 2013 kl. 17:33:20 UTC+1 skrev Craig
> Markwardt:
>> On Wednesday, November 20, 2013 7:53:22 AM UTC-8, fd_...@mail.com
>> wrote:
>>
>>>> Maria, are you going to delete your posts again?
>>
>>> I had delete my post before because instead of singular values I
>>> wrote eigenvalues which is wrong designation. I didn't know how
>>> to edit my post that is why I had delete it and re post it
>>> correctly. I didn't know that we are not allowed to delete our
>>> posts. Sorry about that.
>>
>> Well, you deleted two posts in your "Derivatives" thread, not just
>> one.
>>
>> You can do whatever you want, but if you delete your posts, I'm
>> going to recommend that people don't respond. Deleting one's posts
>> is unfair to other readers and to people that try to respond to the
>> post.
>>
>> Nobody can edit posts. If you make an error - that's fine,
>> everybody does - then just post a follow-up correcting the error.
>
> Deleting posts in google groups is also not very effective. They may
> disappear from google but many proper usenet news servers ignore such
> deletes. Which is another reason to do corrections in followups
> rather than by trying to replace posts with new versions.
Yeah, I was gonna ask about that. I get all the posts (including the
apparently deleted ones).
AFAICT, once you click "send", clicking "delete" is an exercise in
wishful thinking.
Craig needs to use a different news reader/provider/whatevs methinks. :o)
cheers,
paulv
|
|
|
|
|
|
| Re: Determinant of a matrix [message #86587 is a reply to message #86585] |
Thu, 21 November 2013 05:20 |
Paul Van Delst[1]
Messages: 1157
Registered: April 2002
|
Senior Member |
|
|
On 11/20/13 21:33, Craig Markwardt wrote:
> On Wednesday, November 20, 2013 9:04:57 AM UTC-8, Paul van Delst
> wrote:
>> Craig needs to use a different news reader/provider/whatevs
>> methinks. :o)
>
> The servers I've used in the past have all been shut down, so now I
> default to Google Groups, which does honor post cancellations. I
> know it's not "real" but I don't want to spend the time or the money
> to get any "realler." :-)
FWIW, I use nntp.aioe.org through TBird.
It's free, provided you don't post a bzillion messages a day. Good spam
filters too.
cheers,
paulv
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
