Model selection in spline nonparametric regression

A Bayesian approach is presented for model selection in nonparametric regression with Gaussian errors and in binary nonparametric regression. A smoothness prior is assumed for each component of the model and the posterior probabilities of the candidate models are approximated using the Bayesian info...

Full description

Saved in:
Bibliographic Details
Published in:Journal of the Royal Statistical Society. Series B, Statistical methodology Statistical methodology, 2002-01, Vol.64 (1), p.119-139
Main Authors: Wood, Sally, Kohn, Robert, Shively, Tom, Jiang, Wenxin
Format: Article
Language:English
Subjects:
Bayesian analysis
Bayesian information criterion
Bayesian method
Binary regression
Data smoothing
Error
Estimation
Estimators
Exact sciences and technology
Gibbs sampler
Linear inference, regression
Linear regression
Mathematical independent variables
Mathematics
Meteorology
Modeling
Nonparametric inference
Nonparametric models
Ozone
Parametric models
Probabilities
Probability
Probability and statistics
Regression analysis
Sciences and techniques of general use
Statistical methods
Statistical models
Statistics
Thin plate splines
Variable selection
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A Bayesian approach is presented for model selection in nonparametric regression with Gaussian errors and in binary nonparametric regression. A smoothness prior is assumed for each component of the model and the posterior probabilities of the candidate models are approximated using the Bayesian information criterion. We study the model selection method by simulation and show that it has excellent frequentist properties and gives improved estimates of the regression surface. All the computations are carried out efficiently using the Gibbs sampler.
ISSN:1369-7412
1467-9868
DOI:10.1111/1467-9868.00328