Christoph Ruegg

For non-linear regression have a look at the `MathNet.Numerics.Optimization` namespace, which among others also implements gradient-free algorithms like the Nelder-Mead simplex algorithm. See also `FindMinimum` in the root namespace where...

We do accept pull requests - although I've been a bit slow to accept them lately.

I've added a new Fit.Curve function: ```C# /// /// Non-linear least-squares fitting the points (x,y) to an arbitrary function y : x -> f(p, x), /// returning its best fitting...

Impressive, thanks! FYI, currently in mainline in the la-symmetric branch.

Thanks for reporting this!

Yes, OuterProduct needs a special implementation for sparse vectors but does not have one yet.

Thanks! It seems some of the tests actually fail on the build server. Could you have a look? ``` 1) Test Error : MathNet.Numerics.UnitTests.FinancialTests.IndicatorsTests.Atr_ShouldRaiseException_IfPeriodIsGreaterThanBarCount System.FormatException : String was not recognized...

Seems like the dates in the test file are in a culture-local format (German?), without explicitly specifying `de-DE` these tests will fail when run in other cultures.

Thanks for the fix. I've pulled your changes into the atr-indicator branch of this repo. I have a suggestion for an online implementation which could be useful. But not working...

Thanks! I wonder whether we can make it such that any of the [distance functions](http://numerics.mathdotnet.com/Distance.html) can be used here?