Reference Bands for Nonparametrically Estimated Link Functions

Andreas F. Ruckstuh; Alan H. Welsh

doi:10.1080/10618600.1999.10474844

Reference Bands for Nonparametrically Estimated Link Functions

Andreas F. Ruckstuh, Alan H. Welsh

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

5 Citations (Scopus)

Abstract

We explore the shape of the link function in a generalized linear model by estimating the link nonparametrically. We consider the problem of comparing the nonparametrically estimated link with a particular link function. Using reference bands that consist of pointwise confidence intervals for the nonparametrically estimated link centered at the hypothesized parametric link, simple graphical methods are obtained for comparing the two link functions. The resulting diagnostic plots are demonstrated with both artificial and real examples.

Original language	English
Pages (from-to)	699-714
Number of pages	16
Journal	Journal of Computational and Graphical Statistics
Volume	8
Issue number	4
DOIs	https://doi.org/10.1080/10618600.1999.10474844
Publication status	Published - Dec 1999

Access to Document

10.1080/10618600.1999.10474844

Cite this

@article{13c5f439df4a40b984ffe4e07d447b2c,

title = "Reference Bands for Nonparametrically Estimated Link Functions",

abstract = "We explore the shape of the link function in a generalized linear model by estimating the link nonparametrically. We consider the problem of comparing the nonparametrically estimated link with a particular link function. Using reference bands that consist of pointwise confidence intervals for the nonparametrically estimated link centered at the hypothesized parametric link, simple graphical methods are obtained for comparing the two link functions. The resulting diagnostic plots are demonstrated with both artificial and real examples.",

keywords = "Generalized linear model, Graphics, Local polynomial smoothing",

author = "Ruckstuh, {Andreas F.} and Welsh, {Alan H.}",

year = "1999",

month = dec,

doi = "10.1080/10618600.1999.10474844",

language = "English",

volume = "8",

pages = "699--714",

journal = "Journal of Computational and Graphical Statistics",

issn = "1061-8600",

publisher = "Taylor and Francis Ltd.",

number = "4",

}

TY - JOUR

T1 - Reference Bands for Nonparametrically Estimated Link Functions

AU - Ruckstuh, Andreas F.

AU - Welsh, Alan H.

PY - 1999/12

Y1 - 1999/12

N2 - We explore the shape of the link function in a generalized linear model by estimating the link nonparametrically. We consider the problem of comparing the nonparametrically estimated link with a particular link function. Using reference bands that consist of pointwise confidence intervals for the nonparametrically estimated link centered at the hypothesized parametric link, simple graphical methods are obtained for comparing the two link functions. The resulting diagnostic plots are demonstrated with both artificial and real examples.

AB - We explore the shape of the link function in a generalized linear model by estimating the link nonparametrically. We consider the problem of comparing the nonparametrically estimated link with a particular link function. Using reference bands that consist of pointwise confidence intervals for the nonparametrically estimated link centered at the hypothesized parametric link, simple graphical methods are obtained for comparing the two link functions. The resulting diagnostic plots are demonstrated with both artificial and real examples.

KW - Generalized linear model

KW - Graphics

KW - Local polynomial smoothing

UR - http://www.scopus.com/inward/record.url?scp=0033268494&partnerID=8YFLogxK

U2 - 10.1080/10618600.1999.10474844

DO - 10.1080/10618600.1999.10474844

M3 - Article

SN - 1061-8600

VL - 8

SP - 699

EP - 714

JO - Journal of Computational and Graphical Statistics

JF - Journal of Computational and Graphical Statistics

IS - 4

ER -

Reference Bands for Nonparametrically Estimated Link Functions

Abstract

Access to Document

Other files and links

Fingerprint

Cite this