Model selection in spline nonparametric regression

Sally Wood; Robert Kohn; Tom Shively; Wenxin Jiang

doi:10.1111/1467-9868.00328

Model selection in spline nonparametric regression

Sally Wood, Robert Kohn, Tom Shively, Wenxin Jiang

Statistics and Data Science

Research output: Contribution to journal › Article › peer-review

28 Scopus citations

Abstract

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.

Original language	English (US)
Pages (from-to)	119-139
Number of pages	21
Journal	Journal of the Royal Statistical Society. Series B: Statistical Methodology
Volume	64
Issue number	1
DOIs	https://doi.org/10.1111/1467-9868.00328
State	Published - 2002

Keywords

Bayesian analysis
Bayesian information criterion
Binary regression
Gibbs sampler
Thin plate splines
Variable selection

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.1111/1467-9868.00328

Cite this

@article{71591b56d2d64f1b8a6c705e117bc8bf,

title = "Model selection in spline nonparametric regression",

abstract = "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.",

keywords = "Bayesian analysis, Bayesian information criterion, Binary regression, Gibbs sampler, Thin plate splines, Variable selection",

author = "Sally Wood and Robert Kohn and Tom Shively and Wenxin Jiang",

year = "2002",

doi = "10.1111/1467-9868.00328",

language = "English (US)",

volume = "64",

pages = "119--139",

journal = "Journal of the Royal Statistical Society. Series B: Statistical Methodology",

issn = "1369-7412",

publisher = "Wiley-Blackwell",

number = "1",

}

TY - JOUR

T1 - Model selection in spline nonparametric regression

AU - Wood, Sally

AU - Kohn, Robert

AU - Shively, Tom

AU - Jiang, Wenxin

PY - 2002

Y1 - 2002

N2 - 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.

AB - 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.

KW - Bayesian analysis

KW - Bayesian information criterion

KW - Binary regression

KW - Gibbs sampler

KW - Thin plate splines

KW - Variable selection

UR - http://www.scopus.com/inward/record.url?scp=0036427356&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036427356&partnerID=8YFLogxK

U2 - 10.1111/1467-9868.00328

DO - 10.1111/1467-9868.00328

M3 - Article

AN - SCOPUS:0036427356

SN - 1369-7412

VL - 64

SP - 119

EP - 139

JO - Journal of the Royal Statistical Society. Series B: Statistical Methodology

JF - Journal of the Royal Statistical Society. Series B: Statistical Methodology

IS - 1

ER -

Model selection in spline nonparametric regression

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this