Dear all
I found a solution.
I get the column echelon form of A by X=rref(A')'.
Then the rows' numbers with the non-zero pivot of X are the independent
rows' numbers of A.
In this problem 26th row and 27th row are redundant.
26th rows are represented by linear combinations of 24th row and
Dear Wescot
I know how to approximate A with SVD, but I do not know how to find
independent row vectors of A.
How can I do?
Best regards.
--
Sent from:
http://mailinglists.scilab.org/Scilab-users-Mailing-Lists-Archives-f2602246.html
___
users
Sorry, I was werong with the constraint matrix.
The right values are as follows.
A=zeros(27,27)
A(1,10)=1
A(1,10)=1
A(2,5)=1
A(3,14)=1
A(4,23)=1
A(5,7)=1
A(6,16)=1
A(7,25)=1
A(8,17)=1
A(9,1)=0,A(9,18)=1
A(10,2)=1,A(10,3)=1
A(11,4)=1,A(11,5)=1,A(11,6)=1
A(12,7)=1,A(12,8)=1,A(12,9)=1
Is there a reason not to do SVD, and throw out the singular values that
are too small?
On Sun, 2018-12-02 at 09:56 -0700, fujimoto2005 wrote:
> This problem is an economic problem. The i-th row of the square
> constraint
> matrix A with m dimension expresses certain economic constraints.
> The
Hi mottelet.
Thank you for the question.
I'm happy if all possible solutions are available. Since the rank is known
to be 22, the number of solutions is 27Conb22 = 80730 or less, but if the
number is small, I think I can find "basic" constraints that is economically
meaningful.
If it is
Please I want to ask a question. Please I have been trying to export a graphic
from scilab to an MS word document. How do I do that? Please help. Thanks
Sent from Yahoo Mail on Android
On Sun, Dec 2, 2018 at 5:57 PM, fujimoto2005 wrote:
This problem is an economic problem. The i-th row
This problem is an economic problem. The i-th row of the square constraint
matrix A with m dimension expresses certain economic constraints.
The elements of the constraint matrix are either 0 or 1.
Suppose the rank of A is r and by changing the row number a_1, ..., a_r are
linearly independent.
I
Hello
Do you just want one of the solutions, e.g. of minimum norm, or do you want
more precision on the nullspace ?
S.
> Le 2 déc. 2018 à 09:43, fujimoto2005 a écrit :
>
> I am trying to solve linear equations with 30 variables. Since the
> determinant of the coefficient matrix is 0, I can
I am trying to solve linear equations with 30 variables. Since the
determinant of the coefficient matrix is 0, I can tell that some row vectors
are linearly dependent on other row vectors. I want to solve the problem by
deleting the linearly dependent rows while simultaneously changing the
