Dear all,

I am currently experimenting gnumeric's solver.
It is a very useful tool, congratulations!

One feature that I miss, though, is the following.
Often in data inversion problems, one try to minimize a cost function
which is (in the gaussian case) a sum of squared residuals divided by
squared confidence interval.
When reaching the minimum of the cost function, one evaluates the
confidence intervals of the tuning parameters by using the confidence
intervals of the data and by approximating the model by its tangent.

Could this feature be implemented in gnumeric? It would make the
solver a lot more useful.

Thanks and best regards,

Frédéric Parenin



-- 
http://parrenin.frederic.free.fr/
_______________________________________________
gnumeric-list mailing list
[email protected]
https://mail.gnome.org/mailman/listinfo/gnumeric-list

Reply via email to