Re: Noise properties of a linear system if elements are deleted

Peter Spellucci wrote:

you can make this systematically and the approach is known as a "basic least
squares solution" (in contrast to the solution by the pseudoinverse)

Thanks for your help. I have to admit that I don't completely follow
your description however as I understand it, there is no need for prior
knowledge. As I should have pointed out, the system of equations is
frequently under-determined but by performing deletions driven by a
model solution it is possible to make it determined. My feeling is that
with the luxury of prior knwledge it should be possible to do better
than without; is there a standard way of doing this?

.