Computational Statistics

, Volume 15, Issue 4, pp 443–462 | Cite as

A Comparison of Regression Spline Smoothing Procedures

  • M.P. Wand


Regression spline smoothing involves modelling a regression function as a piecewise polynomial with a high number of pieces relative to the sample size. Because the number of possible models is so large, efficient strategies for choosing among them are required. In this paper we review approaches to this problem and compare them through a simulation study. For simplicity and conciseness we restrict attention to the univariate smoothing setting with Gaussian noise and the truncated polynomial regression spline basis.

Keywords: Bayesian variable selection, B-spline, Gibbs sampling, Non-parametric regression, Polynomial spline, Roughness penalty, Stepwise regression. 

Copyright information

© Physica-Verlag, Heidelberg 2000

Authors and Affiliations

  • M.P. Wand
    • 1
  1. 1.Department of Biostatistics, School of Public Health, Harvard University, 665 Huntington Avenue, Boston, MA 02115, USAUS

Personalised recommendations