Splice plots for generalized linear models

David Nelson; Siamak Noorbaloochi

doi:10.1080/03610926.2014.895842

Splice plots for generalized linear models

David Nelson, Siamak Noorbaloochi

Medicine - Veteran's Administration Medical Center

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

Abstract

ABSTRACT: We develop splice plots as a diagnostic tool for parametric generalized linear models. Splice plots use the independence of the outcome and explanatory measures given the regression function. Plotting differences between the estimated parametric regression function and non-parametric estimates of the regression function computed in small neighborhoods of the fitted values from the parametric model can be used to assess model fit.

Original language	English (US)
Pages (from-to)	3524-3540
Number of pages	17
Journal	Communications in Statistics - Theory and Methods
Volume	45
Issue number	12
DOIs	https://doi.org/10.1080/03610926.2014.895842
State	Published - Jun 17 2016

Bibliographical note

Publisher Copyright:
© 2016 Taylor & Francis Group, LLC.

Keywords

Model diagnostics
Model selection
Sufficient summary

Access

10.1080/03610926.2014.895842

OpenUrl availability

Full text

Cite this

@article{632e6c83611c48c898c3d6d9de805dc3,

title = "Splice plots for generalized linear models",

abstract = "ABSTRACT: We develop splice plots as a diagnostic tool for parametric generalized linear models. Splice plots use the independence of the outcome and explanatory measures given the regression function. Plotting differences between the estimated parametric regression function and non-parametric estimates of the regression function computed in small neighborhoods of the fitted values from the parametric model can be used to assess model fit.",

keywords = "Model diagnostics, Model selection, Sufficient summary",

author = "David Nelson and Siamak Noorbaloochi",

note = "Publisher Copyright: {\textcopyright} 2016 Taylor & Francis Group, LLC.",

year = "2016",

month = jun,

day = "17",

doi = "10.1080/03610926.2014.895842",

language = "English (US)",

volume = "45",

pages = "3524--3540",

journal = "Communications in Statistics - Theory and Methods",

issn = "0361-0926",

publisher = "Taylor and Francis Ltd.",

number = "12",

}

TY - JOUR

T1 - Splice plots for generalized linear models

AU - Nelson, David

AU - Noorbaloochi, Siamak

PY - 2016/6/17

Y1 - 2016/6/17

N2 - ABSTRACT: We develop splice plots as a diagnostic tool for parametric generalized linear models. Splice plots use the independence of the outcome and explanatory measures given the regression function. Plotting differences between the estimated parametric regression function and non-parametric estimates of the regression function computed in small neighborhoods of the fitted values from the parametric model can be used to assess model fit.

AB - ABSTRACT: We develop splice plots as a diagnostic tool for parametric generalized linear models. Splice plots use the independence of the outcome and explanatory measures given the regression function. Plotting differences between the estimated parametric regression function and non-parametric estimates of the regression function computed in small neighborhoods of the fitted values from the parametric model can be used to assess model fit.

KW - Model diagnostics

KW - Model selection

KW - Sufficient summary

UR - http://www.scopus.com/inward/record.url?scp=84970023337&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84970023337&partnerID=8YFLogxK

U2 - 10.1080/03610926.2014.895842

DO - 10.1080/03610926.2014.895842

M3 - Article

AN - SCOPUS:84970023337

SN - 0361-0926

VL - 45

SP - 3524

EP - 3540

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

IS - 12

ER -

Splice plots for generalized linear models

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this