[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Retract [Bug-gsl] gsl_multifit_linear() covariances incorrect due to
Re: Retract [Bug-gsl] gsl_multifit_linear() covariances incorrect due to sorted singular values?
Fri, 09 Feb 2007 14:27:04 +0000
Wanderlust/2.14.0 (Africa) Emacs/21.3 Mule/5.0 (SAKAKI)
At Thu, 08 Feb 2007 14:56:47 -0500,
Doug Rogers wrote:
> Doug Rogers wrote:
> > I've noticed that the singular values returned by SVD are sorted,
> > destroying the relationship that they had with their column vector. This
> > means, I believe, that the covariances calculated in
> > gsl_multifit_wlinear(), et al., are not in the proper order - they
> > correspond to different solution coefficients.
> Well, it looks like it was operator error. I summed over the wrong
> index. Sorry for any pain I may have caused.
Glad to hear you got it sorted out.
Information about GSL development can be found at
http://www.gnu.org/software/gsl/#volunteer ("How to help")
Network Theory Ltd,
Publishing the GSL Manual - http://www.network-theory.co.uk/gsl/manual/