NMath currently supports basic PLS regression, with cross-validation. It would not be difficult to implement these other measures within NMath, but they are not currently on our product roadmap. You could either implement them yourself, or we could do it on a consulting basis.

Something we’ve been doing a lot lately is “partnered development.” If a customer really needs a feature, and it is a good fit for the library, we will do the work on a contract basis at a reduced fee. This works well for both the customer and us – they get the work done for about the same cost, or less, than if they had to do it themselves (plus they don’t have to worry about maintaining the code), and we get added functionality to our library for a relatively low cost.

If you’re interesting in discussing partnered development, please contact sales@centerspace.net.

Thanks,
Ken

This version is nice to use. For PLS regression, is it possible to support Q^2, R2X, R2Y, and VIP (Variable Important in Projection)?