Nonparametric kernel regression with multiple predictors and multiple shape constraints

Pang Du, Christopher Parmeter, Jeffrey S. Racine

Research output: Contribution to journalArticle

46 Citations (Scopus)

Abstract

Nonparametric smoothing under shape constraints has recently received much well-deserved attention. Powerful methods have been proposed for imposing a single shape constraint such as monotonicity and concavity on univariate functions. In this paper, we extend the monotone kernel regression method in Hall and Huang (2001) to the multivariate and multi-constraint setting. We impose equality and/or inequality constraints on a nonparametric kernel regression model and its derivatives. A bootstrap procedure is also proposed for testing the validity of the constraints. Consistency of our constrained kernel estimator is provided through an asymptotic analysis of its relationship with the unconstrained estimator. Theoretical underpinnings for the bootstrap procedure are also provided. Illustrative Monte Carlo results are presented and an application is considered.

Original languageEnglish (US)
Pages (from-to)1347-1371
Number of pages25
JournalStatistica Sinica
Volume23
Issue number3
DOIs
StatePublished - Jul 1 2013

Fingerprint

Shape Constraint
Kernel Regression
Nonparametric Regression
Bootstrap
Predictors
Monotone Regression
Nonparametric Smoothing
Kernel Estimator
Concavity
Inequality Constraints
Asymptotic Analysis
Univariate
Monotonicity
Regression Model
Equality
Estimator
Derivative
Testing
Kernel regression
Relationships

Keywords

  • Hypothesis testing
  • Multivariate kernel estimation
  • Nonparametric regression
  • Shape restrictions

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Nonparametric kernel regression with multiple predictors and multiple shape constraints. / Du, Pang; Parmeter, Christopher; Racine, Jeffrey S.

In: Statistica Sinica, Vol. 23, No. 3, 01.07.2013, p. 1347-1371.

Research output: Contribution to journalArticle

@article{1e74dfadc3db41b1a991b5ffb225b957,
title = "Nonparametric kernel regression with multiple predictors and multiple shape constraints",
abstract = "Nonparametric smoothing under shape constraints has recently received much well-deserved attention. Powerful methods have been proposed for imposing a single shape constraint such as monotonicity and concavity on univariate functions. In this paper, we extend the monotone kernel regression method in Hall and Huang (2001) to the multivariate and multi-constraint setting. We impose equality and/or inequality constraints on a nonparametric kernel regression model and its derivatives. A bootstrap procedure is also proposed for testing the validity of the constraints. Consistency of our constrained kernel estimator is provided through an asymptotic analysis of its relationship with the unconstrained estimator. Theoretical underpinnings for the bootstrap procedure are also provided. Illustrative Monte Carlo results are presented and an application is considered.",
keywords = "Hypothesis testing, Multivariate kernel estimation, Nonparametric regression, Shape restrictions",
author = "Pang Du and Christopher Parmeter and Racine, {Jeffrey S.}",
year = "2013",
month = "7",
day = "1",
doi = "10.5705/ss.2012.024",
language = "English (US)",
volume = "23",
pages = "1347--1371",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Institute of Statistical Science",
number = "3",

}

TY - JOUR

T1 - Nonparametric kernel regression with multiple predictors and multiple shape constraints

AU - Du, Pang

AU - Parmeter, Christopher

AU - Racine, Jeffrey S.

PY - 2013/7/1

Y1 - 2013/7/1

N2 - Nonparametric smoothing under shape constraints has recently received much well-deserved attention. Powerful methods have been proposed for imposing a single shape constraint such as monotonicity and concavity on univariate functions. In this paper, we extend the monotone kernel regression method in Hall and Huang (2001) to the multivariate and multi-constraint setting. We impose equality and/or inequality constraints on a nonparametric kernel regression model and its derivatives. A bootstrap procedure is also proposed for testing the validity of the constraints. Consistency of our constrained kernel estimator is provided through an asymptotic analysis of its relationship with the unconstrained estimator. Theoretical underpinnings for the bootstrap procedure are also provided. Illustrative Monte Carlo results are presented and an application is considered.

AB - Nonparametric smoothing under shape constraints has recently received much well-deserved attention. Powerful methods have been proposed for imposing a single shape constraint such as monotonicity and concavity on univariate functions. In this paper, we extend the monotone kernel regression method in Hall and Huang (2001) to the multivariate and multi-constraint setting. We impose equality and/or inequality constraints on a nonparametric kernel regression model and its derivatives. A bootstrap procedure is also proposed for testing the validity of the constraints. Consistency of our constrained kernel estimator is provided through an asymptotic analysis of its relationship with the unconstrained estimator. Theoretical underpinnings for the bootstrap procedure are also provided. Illustrative Monte Carlo results are presented and an application is considered.

KW - Hypothesis testing

KW - Multivariate kernel estimation

KW - Nonparametric regression

KW - Shape restrictions

UR - http://www.scopus.com/inward/record.url?scp=84883879522&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883879522&partnerID=8YFLogxK

U2 - 10.5705/ss.2012.024

DO - 10.5705/ss.2012.024

M3 - Article

AN - SCOPUS:84883879522

VL - 23

SP - 1347

EP - 1371

JO - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 3

ER -