Adaptive spline fitting with particle swarm optimization
From MaRDI portal
Abstract: In fitting data with a spline, finding the optimal placement of knots can significantly improve the quality of the fit. However, the challenging high-dimensional and non-convex optimization problem associated with completely free knot placement has been a major roadblock in using this approach. We present a method that uses particle swarm optimization (PSO) combined with model selection to address this challenge. The problem of overfitting due to knot clustering that accompanies free knot placement is mitigated in this method by explicit regularization, resulting in a significantly improved performance on highly noisy data. The principal design choices available in the method are delineated and a statistically rigorous study of their effect on performance is carried out using simulated data and a wide variety of benchmark functions. Our results demonstrate that PSO-based free knot placement leads to a viable and flexible adaptive spline fitting approach that allows the fitting of both smooth and non-smooth functions.
Recommendations
Cites work
- scientific article; zbMATH DE number 45848 (Why is no real title available?)
- scientific article; zbMATH DE number 48302 (Why is no real title available?)
- scientific article; zbMATH DE number 1131224 (Why is no real title available?)
- scientific article; zbMATH DE number 1917660 (Why is no real title available?)
- A Data-Reduction Strategy for Splines with Applications to the Approximation of Functions and Data
- A comparison of regression spline smoothing procedures
- A practical guide to splines.
- Adapting to Unknown Smoothness via Wavelet Shrinkage
- Adaptive penalized splines for data smoothing
- Approximation to Data by Splines with Free Knots
- Asymptotic properties of penalized spline estimators
- Automatic Bayesian Curve Fitting
- Bayesian curve-fitting with free-knot splines
- Data driven adaptive spline smoothing
- Flexible Parsimonious Smoothing and Additive Modeling
- Flexible smoothing with B-splines and penalties. With comments and a rejoinder by the authors
- Functional data analysis
- Generalized Cross-Validation as a Method for Choosing a Good Ridge Parameter
- Hybrid Adaptive Splines
- Knot calculation for spline fitting based on the unimodality property
- Locally adaptive Bayesian P-splines with a normal-exponential-gamma prior
- Minimization by Random Search Techniques
- Multivariate adaptive regression splines
- On calculating with B-splines
- Polynomial splines and their tensor products in extended linear modeling. (With discussions)
- Reversible jump Markov chain Monte Carlo computation and Bayesian model determination
- Semiparametric Regression
- Smoothing parameter selection in two frameworks for penalized splines
- Smoothing by spline functions.
- Smoothing noisy data with spline functions: Estimating the correct degree of smoothing by the method of generalized cross-validation
- Smoothing splines with varying smoothing parameter
- Spatially Adaptive Regression Splines and Accurate Knot Selection Schemes
- Spline Functions in Data Analysis
- Splines (with optimal knots) are better
- Splines in Statistics
- Swarm intelligence methods for statistical regression
- Twenty years of P-splines (invited article)
Cited in
(2)
This page was built for publication: Adaptive spline fitting with particle swarm optimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1995828)
