scientific article; zbMATH DE number 19409
[For the entire collection see Zbl 0733.00012.] The author deals with the computation of the coefficients \((Y^ n_ j)\) of a cubic spline series \(\sigma^ n(x)=\sum_ jY_ j^ nB_ j(x)\) (defined with respect to equally spaced breakpoints \(x_ j=jh\) of a given step size \(h)\) as linear combinations of given data \((y_ j)\). Several ways are suggested to fixe the \(2n+1\) coefficients of these linear combinations using the socalled transfer function \(H_ n\) of the associated linear transformation \(T_ n:(y_ j)\in\mathbb{R}^ \mathbb{Z}\to(\sigma^ n(x_ j))\in\mathbb{R}^ \mathbb{Z}\), where for \(\alpha\in\mathbb{R}\), \(H_ n(\alpha)\) is the eigenvalue belonging to the eigenvector \(c_ \alpha=(e^{2\pi i\alpha k})_{k\in\mathbb{Z}}\) of \(T_ n\). The use of this transfer function is discussed, including some examples.
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 Q3978895)
