Displaying 1 result from an estimated 1 matches for "wsqrt".
Did you mean:
sqrt
2001 Nov 20
0
Summary: non-negative least squares
...nd the scaling is often a workaround.
I don't know how this compares to Brian Ripleys suggestion (use
optim) in times of speed and accuracy, but it should be competitive.
Yours,
Marcel
#---------------------------------------------------------------------------
--
nnls.fit <- function(x,y,wsqrt=1,eps=0,rank.tol=1e-07) {
## Purpose: Nonnegative Least Squares (similar to the S-Plus function
## with the same name) with the help of the R-library quadprog
## ------------------------------------------------------------------------
## Attention:
## - weights are square roots of usual w...