[Scilab-users] datafit.sci with W

[Samuel Gougeon]

Hello,

Le 19/09/2012 16:48, RolandKoberle a écrit :
> .../...
> The help page says:
> W:    weighting matrix of size ne x ne (optional; defaut no ponderation)
>
> But if W weighs the data-points ( their number is nz ) , then W should 
> be a vector of length nz?
>
> If you try to run datafit with W = eye(ne,ne) or W=ones(nz,1) etc., 
> you get an error message ?????
W does not weight data but criteria. It has been discussed in this thread:
http://mailinglists.scilab.org/Weighting-data-in-datafit-tt3633921.html#a3636450

Regards
Samuel