Universally consistent regression function estimation using hierarchical B-splines

DOI10.1006/JMVA.1998.1782MaRDI QIDQ1283852zbMATH OpenOpenAlexFDO

Authors Michael Kohler

Publication date 30 March 1999

Published in Journal of Multivariate Analysis (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1006/jmva.1998.1782

zbMATH Keywords

data-dependent partitionsintegrated squared errorleast squares estimateregression estimateuniversal consistency

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Asymptotic properties of nonparametric inference (62G20) Estimation in multivariate analysis (62H12) Numerical computation using splines (65D07)

Review text

Estimation of multivariate regression functions from i.i.d. data is considered. By empirical \(L_2\)-error minimization over data-dependent spaces of polynomial spline functions the estimates have been constructed. For univariate regression function estimation these spaces are spline-spaces with data-dependent knot sequences. In the multivariate case, the author uses so-called hierarchical spline spaces which are defined as a linear span of tensor product \(B\)-splines with nested knot sequences. This paper shows the strong \(L_2\)-consistency of the estimators without any condition on the underlying distribution.

Recommendations

• On the universal consistency of a least squares spline regression estimator
• Universal consistency of local polynomial kernel regression estimates
• The use of polynomial splines and their tensor products in multivariate function estimation. (With discussion)
• scientific article; zbMATH DE number 17215

Cites work

• A practical guide to splines
• Central limit theorems for empirical measures
• Consistency of data-driven histogram methods for density estimation and classification
• Consistent nonparametric regression. Discussion
• Convergence of stochastic processes
• Decision theoretic generalizations of the PAC model for neural net and other learning applications
• Histogram regression estimation using data-dependent partitions
• scientific article; zbMATH DE number 3703310 (Why is no real title available?)
• scientific article; zbMATH DE number 1136506 (Why is no real title available?)
• scientific article; zbMATH DE number 774848 (Why is no real title available?)
• scientific article; zbMATH DE number 893887 (Why is no real title available?)
• scientific article; zbMATH DE number 1404604 (Why is no real title available?)
• Nonparametric estimation via empirical risk minimization
• On the strong universal consistency of a recursive regression estimate by Pál Révész
• On the strong universal consistency of a series type regression estimate
• On the strong universal consistency of nearest neighbor regression function estimates
• On the universal consistency of a least squares spline regression estimator
• The use of polynomial splines and their tensor products in multivariate function estimation. (With discussion)

Cited in

(10)

• On the universal consistency of a least squares spline regression estimator
• Prediction from randomly right censored data
• On the strong universal consistency of local averaging regression estimates
• Rates of convergence for partitioning and nearest neighbor regression estimates with unbounded data
• Strong universal consistency of smooth kernel regression estimates
• Nonparametric regression with additional measurement errors in the dependent variable
• Rates of convergence for the \(k\)-nearest neighbor estimators with smoother regression functions
• Nonparametric regression function estimation using interaction least squares splines and complexity regularization.
• Strongly consistent density estimation of the regression residual
• Optimal global rates of convergence for nonparametric regression with unbounded data

This page was built for publication: Universally consistent regression function estimation using hierarchical \(B\)-splines

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1283852)