[Quantlib-dev] NonLeastSquare with LevenbergMarquardt: small code update

[Yomi <gui...@gm...>]

Hi,

I was trying to make the NonLeastSquare algorithm.
The results were not performing well using ConjugateGradient, so I have
decided to use LevenbergMarquardt instead.

To make it work, you need to update the LeastSquareFunction values member
function as follow:

Disposable<Array> LeastSquareFunction::values(const Array& x) const {
	
            // size of target and function to fit vectors
            Array target(lsp_.size()), fct2fit(lsp_.size());
           
           // compute its values
           lsp_.targetAndValue(x, target, fct2fit);
           
           // do the difference
           Array diff = (target - fct2fit)*(target - fct2fit);

           return	diff;
           }

Of course don't forget to remove the definition from the class declaration.
After this, the results are very close from what I can get elsewhere.

Cheers
-- 
View this message in context: http://www.nabble.com/NonLeastSquare-with-LevenbergMarquardt%3A-small-code-update-tf4730027.html#a13525064
Sent from the quantlib-dev mailing list archive at Nabble.com.