Using P-splines to test the linearity of partially linear models

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

The nonparametric component in a partially linear model is estimated by a linear combination of fixed-knot cubic B-splines with a second-order difference penalty on the adjacent B-spline coefficients. The resulting penalized least-squares estimator is used to construct two Wald-type spline-based test statistics for the null hypothesis of the linearity of the nonparametric function. When the number of knots is fixed, the first test statistic asymptotically has the distribution of a linear combination of independent chi-squared random variables, each with one degree of freedom, under the null hypothesis. The smoothing parameter is determined by specifying a value for the asymptotically expected value of the test statistic under the null hypothesis. When the number of knots is fixed and under the null hypothesis, the second test statistic asymptotically has a chi-squared distribution with K = q + 2 degrees of freedom, where q is the number of knots used for estimation. The power performances of the two proposed tests are investigated via simulation experiments, and the practicality of the proposed methodology is illustrated using a real-life data set.

Original languageEnglish (US)
Pages (from-to)542-552
Number of pages11
JournalStatistical Methodology
Volume6
Issue number5
DOIs
StatePublished - Sep 2009

Fingerprint

P-splines
Partially Linear Model
Linearity
Null hypothesis
Knot
Test Statistic
Linear Combination
Degree of freedom
Penalized Least Squares
Cubic B-spline
Chi-squared distribution
Chi-squared
Smoothing Parameter
Least Squares Estimator
B-spline
Expected Value
Simulation Experiment
Spline
Penalty
Adjacent

Keywords

  • B-splines
  • P-splines
  • Penalized least-squares
  • Wald-type spline-based test

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Using P-splines to test the linearity of partially linear models. / Li, Chin-Shang.

In: Statistical Methodology, Vol. 6, No. 5, 09.2009, p. 542-552.

Research output: Contribution to journalArticle

@article{c921702bc0094ac195c9584f34b73061,
title = "Using P-splines to test the linearity of partially linear models",
abstract = "The nonparametric component in a partially linear model is estimated by a linear combination of fixed-knot cubic B-splines with a second-order difference penalty on the adjacent B-spline coefficients. The resulting penalized least-squares estimator is used to construct two Wald-type spline-based test statistics for the null hypothesis of the linearity of the nonparametric function. When the number of knots is fixed, the first test statistic asymptotically has the distribution of a linear combination of independent chi-squared random variables, each with one degree of freedom, under the null hypothesis. The smoothing parameter is determined by specifying a value for the asymptotically expected value of the test statistic under the null hypothesis. When the number of knots is fixed and under the null hypothesis, the second test statistic asymptotically has a chi-squared distribution with K = q + 2 degrees of freedom, where q is the number of knots used for estimation. The power performances of the two proposed tests are investigated via simulation experiments, and the practicality of the proposed methodology is illustrated using a real-life data set.",
keywords = "B-splines, P-splines, Penalized least-squares, Wald-type spline-based test",
author = "Chin-Shang Li",
year = "2009",
month = "9",
doi = "10.1016/j.stamet.2009.06.001",
language = "English (US)",
volume = "6",
pages = "542--552",
journal = "Statistical Methodology",
issn = "1572-3127",
publisher = "Elsevier",
number = "5",

}

TY - JOUR

T1 - Using P-splines to test the linearity of partially linear models

AU - Li, Chin-Shang

PY - 2009/9

Y1 - 2009/9

N2 - The nonparametric component in a partially linear model is estimated by a linear combination of fixed-knot cubic B-splines with a second-order difference penalty on the adjacent B-spline coefficients. The resulting penalized least-squares estimator is used to construct two Wald-type spline-based test statistics for the null hypothesis of the linearity of the nonparametric function. When the number of knots is fixed, the first test statistic asymptotically has the distribution of a linear combination of independent chi-squared random variables, each with one degree of freedom, under the null hypothesis. The smoothing parameter is determined by specifying a value for the asymptotically expected value of the test statistic under the null hypothesis. When the number of knots is fixed and under the null hypothesis, the second test statistic asymptotically has a chi-squared distribution with K = q + 2 degrees of freedom, where q is the number of knots used for estimation. The power performances of the two proposed tests are investigated via simulation experiments, and the practicality of the proposed methodology is illustrated using a real-life data set.

AB - The nonparametric component in a partially linear model is estimated by a linear combination of fixed-knot cubic B-splines with a second-order difference penalty on the adjacent B-spline coefficients. The resulting penalized least-squares estimator is used to construct two Wald-type spline-based test statistics for the null hypothesis of the linearity of the nonparametric function. When the number of knots is fixed, the first test statistic asymptotically has the distribution of a linear combination of independent chi-squared random variables, each with one degree of freedom, under the null hypothesis. The smoothing parameter is determined by specifying a value for the asymptotically expected value of the test statistic under the null hypothesis. When the number of knots is fixed and under the null hypothesis, the second test statistic asymptotically has a chi-squared distribution with K = q + 2 degrees of freedom, where q is the number of knots used for estimation. The power performances of the two proposed tests are investigated via simulation experiments, and the practicality of the proposed methodology is illustrated using a real-life data set.

KW - B-splines

KW - P-splines

KW - Penalized least-squares

KW - Wald-type spline-based test

UR - http://www.scopus.com/inward/record.url?scp=69649093729&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69649093729&partnerID=8YFLogxK

U2 - 10.1016/j.stamet.2009.06.001

DO - 10.1016/j.stamet.2009.06.001

M3 - Article

AN - SCOPUS:69649093729

VL - 6

SP - 542

EP - 552

JO - Statistical Methodology

JF - Statistical Methodology

SN - 1572-3127

IS - 5

ER -