forum.alglib.net

Hi

I'm trying to implement a boundary constrained Linear Least Squares problem using ALGLIB.

To put the problem simply, I want to find the matrix W from:

X * W = T

where
X is a matrix with 50 columns and 3 rows (3 x 50),
W is a column matrix with 50 elements (50 x 1) and
T is a column matrix with 3 element (50 x 1)

The constraints are on the values of the elements of W - let's call the elements 'w'.
0 <= w <= 1.

I've looked at the lsfitlinearc function but there doesn't seem to be an option to include boundary constraints. I also looked at the lsfit_d_nlfb example but I can't seem to comprehend how I could use it for my problem *embarassed*.

Could someone kindly help me out?
Thanks!

You may reformulate it as general optimization problem and solve with BLEIC optimizer. It may take some time, but it should converge unless you have very badly conditioned problem.

Thanks for your reply, I solved it yesterday doing the same and using lsfitfit()