Great! I finally got it sir. Please, I would like to know why we used this orthonormal/orthogonal concenpt in the Beveridge Nelson decomposition. Is it to preserved the order of the alpha or beta matrix, or why exactly do we employ it sir?
Thanks. 

Regards 
Timmy 

On Fri, Mar 15, 2019, 1:37 PM Sven Schreiber <svetosch@gmx.net> wrote:
Am 15.03.2019 um 12:35 schrieb Olasehinde Timmy:

> illustrative purposes using vecm output, let α = {-0.3;0.4}, and β =
> {1,2.2}, how can I compute the orthonormals for this two matrices.
> Because, I tried using the nullspace command in gretl, but I keep
> getting empty space matrix.

You have to use the transpose:

a = {-0.3;0.4}
eval nullspace(a')

This is due to the convention used that A*nullspace(A) should be zero,
as opposed to A'nullspace(A). Equivalently as stated in the
documentation: "...except when A has full column rank, in which case an
empty matrix is returned."

(BTW, the most common name for this is "orthogonal complement" AFAIK.)

cheers
sven
_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users