A spline-based lack-of-fit test for independent variable effect in poisson regression

Chin-Shang Li, Wanzhu Tu

Research output: Contribution to journalArticle

Abstract

In regression analysis of count data, independent variables are often modeled by their linear effects under the assumption of log-linearity. In reality, the validity of such an assumption is rarely tested, and its use is at times unjustifiable. A lack-of-fit test is proposed for the adequacy of a postulated functional form of an independent variable within the framework of semiparametric Poisson regression models based on penalized splines. It offers added flexibility in accommodating the potentially non-loglinear effect of the independent variable. A likelihood ratio test is constructed for the adequacy of the postulated parametric form, for example log-linearity, of the independent variable effect. Simulations indicate that the proposed model performs well, and misspecified parametric model has much reduced power. An example is given.

Original languageEnglish (US)
Pages (from-to)239-247
Number of pages9
JournalJournal of Modern Applied Statistical Methods
Volume6
Issue number1
StatePublished - May 2007
Externally publishedYes

Fingerprint

Lack-of-fit Test
Poisson Regression
Spline
Linearity
Misspecified Model
Semiparametric Regression
Penalized Splines
Count Data
Poisson Model
Likelihood Ratio Test
Parametric Model
Regression Analysis
Regression Model
Flexibility
Model-based
Poisson regression
Splines
Simulation

Keywords

  • B-splines
  • Likelihood ratio test
  • Loglinear model
  • Penalized likelihood
  • Poisson regression model

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Statistics and Probability

Cite this

A spline-based lack-of-fit test for independent variable effect in poisson regression. / Li, Chin-Shang; Tu, Wanzhu.

In: Journal of Modern Applied Statistical Methods, Vol. 6, No. 1, 05.2007, p. 239-247.

Research output: Contribution to journalArticle

@article{70751e60da8e4696a8d5f44477631b31,
title = "A spline-based lack-of-fit test for independent variable effect in poisson regression",
abstract = "In regression analysis of count data, independent variables are often modeled by their linear effects under the assumption of log-linearity. In reality, the validity of such an assumption is rarely tested, and its use is at times unjustifiable. A lack-of-fit test is proposed for the adequacy of a postulated functional form of an independent variable within the framework of semiparametric Poisson regression models based on penalized splines. It offers added flexibility in accommodating the potentially non-loglinear effect of the independent variable. A likelihood ratio test is constructed for the adequacy of the postulated parametric form, for example log-linearity, of the independent variable effect. Simulations indicate that the proposed model performs well, and misspecified parametric model has much reduced power. An example is given.",
keywords = "B-splines, Likelihood ratio test, Loglinear model, Penalized likelihood, Poisson regression model",
author = "Chin-Shang Li and Wanzhu Tu",
year = "2007",
month = "5",
language = "English (US)",
volume = "6",
pages = "239--247",
journal = "Journal of Modern Applied Statistical Methods",
issn = "1538-9472",
publisher = "Wayne State University",
number = "1",

}

TY - JOUR

T1 - A spline-based lack-of-fit test for independent variable effect in poisson regression

AU - Li, Chin-Shang

AU - Tu, Wanzhu

PY - 2007/5

Y1 - 2007/5

N2 - In regression analysis of count data, independent variables are often modeled by their linear effects under the assumption of log-linearity. In reality, the validity of such an assumption is rarely tested, and its use is at times unjustifiable. A lack-of-fit test is proposed for the adequacy of a postulated functional form of an independent variable within the framework of semiparametric Poisson regression models based on penalized splines. It offers added flexibility in accommodating the potentially non-loglinear effect of the independent variable. A likelihood ratio test is constructed for the adequacy of the postulated parametric form, for example log-linearity, of the independent variable effect. Simulations indicate that the proposed model performs well, and misspecified parametric model has much reduced power. An example is given.

AB - In regression analysis of count data, independent variables are often modeled by their linear effects under the assumption of log-linearity. In reality, the validity of such an assumption is rarely tested, and its use is at times unjustifiable. A lack-of-fit test is proposed for the adequacy of a postulated functional form of an independent variable within the framework of semiparametric Poisson regression models based on penalized splines. It offers added flexibility in accommodating the potentially non-loglinear effect of the independent variable. A likelihood ratio test is constructed for the adequacy of the postulated parametric form, for example log-linearity, of the independent variable effect. Simulations indicate that the proposed model performs well, and misspecified parametric model has much reduced power. An example is given.

KW - B-splines

KW - Likelihood ratio test

KW - Loglinear model

KW - Penalized likelihood

KW - Poisson regression model

UR - http://www.scopus.com/inward/record.url?scp=48249115994&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=48249115994&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:48249115994

VL - 6

SP - 239

EP - 247

JO - Journal of Modern Applied Statistical Methods

JF - Journal of Modern Applied Statistical Methods

SN - 1538-9472

IS - 1

ER -