EViews.com

Posted: **Fri Jun 14, 2013 2:28 am**

Hi,

For a given (N x M) matrix A, the theorem on singular value decomposition states that:

A = U * S * V

where the columns of U are the eigenvectors of AA' and the columns of V are eigenvectors of A'A. And S is a diagonal matrix containing the square roots of non-negative eigenvalues from U in descending order.

In Eviews, the command @svd(A) returns the U matrix for A. I was wondering if there is a way to directly get the diagonal matrix S. Can anyone help?

Thanks.

Posted: **Fri Jun 14, 2013 3:12 am**

Code: Select all

'create a random matrix
matrix A = @mnrnd(4,3)

'define the components
matrix U
vector S
matrix V

'decompose the matrix A
U = @svd(A,S,V)

'if you like S as a diagonal matrix
matrix MS = @makediagonal(S)

Posted: **Fri Jun 14, 2013 5:56 am**

Thank you very much! It was very helpful.

Posted: **Fri May 05, 2017 10:17 am**

I get the "mistmatch message" when using @svd to a non-symetric matrix of 37x51. Are u sure, the formula can run for non-symetric matrices?

EViews.com

Singular Value Decomposition (@svd)

Singular Value Decomposition (@svd)

Re: Singular Value Decomposition (@svd)

Re: Singular Value Decomposition (@svd)

Re: Singular Value Decomposition (@svd)