Statistical methods in optimal curve fitting |
| |
| Authors: | Howard L. Weinert |
| |
| Affiliation: | Department of Electrical Engineering , The Johns Hopkins University , Baltimore, Maryland |
| |
| Abstract: | Many optimal curve fitting and approximation problems have the same structure as certain estimation problems involving random processes. This structural correspondence has many useful consequences for curve fitting problems, including recursive algorithms and computable error bounds. The basic facts of this correspondence are reviewed and some new results on error bounds and optimal sampling are presented. |
| |
| Keywords: | splines reproducing kernels interpolation smoothing least-squares estimation |
|
|
