Re: Jacobian of exponential chi2 function -- August 28, 2014 - 14:29

Do you have a reference for a chi-squared Jacobian?  I didn't know
that such a thing existed.  As far as I know, the value, gradient, and
Hessian are at the chi-squared function level.  Whereas the Jacobian
is the matrix of first partial derivatives prior to the sum of
squares.  I.e. in this case, the partial derivatives of the
exponential functions.

If it does exist, could you derive the equations required?  We might
see that that actually explains the factor of two :)

Cheers,

Edward


On 28 August 2014 14:16, Troels Emtekær Linnet <tlinnet@xxxxxxxxx> wrote:
> Hi Edward.
>
> Could you maybe implement the Jacobian from the chi2 exponential function,
> and return as list of lists?
>
> Covar could either be equal
>
> Qxx = ( J^T W J) - 1
> or
> Qxx = ( Jchi2^T Jchi2)-1
> Maybe some scaling on the last
>
> Best
> Troels
> _______________________________________________
> relax (http://www.nmr-relax.com)
>
> This is the relax-devel mailing list
> relax-devel@xxxxxxx
>
> To unsubscribe from this list, get a password
> reminder, or change your subscription options,
> visit the list information page at
> https://mail.gna.org/listinfo/relax-devel