scientific article; zbMATH DE number 3934251

From MaRDI portal
Publication:3706364





[For the entire collection see Zbl 0564.00018.] We are interested in the Hilbert space version of constrained ridge regression, which we will show has many interesting applications. In this paper we propose an extension of the generalized cross validation (GCV) method, to the constrained case. The GCV estimate of \(\lambda\) we propose in the constrained case can be expensive to compute. Thus we propose a first order approximation to it which is very much cheaper to compute, and appears to be satisfactory in the examples we tried.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3706364)