Generalized cross-validation for large scale problems
Proceedings of the second international workshop on Recent advances in total least squares techniques and errors-in-variables modeling
Rank-deficient and discrete ill-posed problems: numerical aspects of linear inversion
Rank-deficient and discrete ill-posed problems: numerical aspects of linear inversion
Smoothing parameter selection for smoothing splines: a simulation study
Computational Statistics & Data Analysis
Semiparametric stochastic frontier models for clustered data
Computational Statistics & Data Analysis
| Hi-index | 0.03 |
A common frame of template splines that unifies the definitions of various spline families, such as smoothing, regression or penalized splines, is considered. The nonlinear nonparametric regression problem that defines the template splines can be reduced, for a large class of Hilbert spaces, to a parameterized regularized linear least squares problem, which leads to an important computational advantage. Particular applications of template splines include the commonly used types of splines, as well as other atypical formulations. In particular, this extension allows an easy incorporation of additional constraints, which is generally not possible in the context of classical spline families.