Assessing influence on the goodness-of-link in generalized linear models

The evidence for a hypothesized link function in a generalized linear model may depend crucially on one or a few observations. Global and local influence diagnostics are proposed to assess influence on the goodness-of-link test. It is found that the local influence method can reveal influential obse...

Full description

Saved in:
Bibliographic Details
Published inStatistics & probability letters Vol. 31; no. 4; pp. 351 - 358
Main Authors Lee, Andy H., Zhao, Yuejen
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.02.1997
Elsevier
SeriesStatistics & Probability Letters
Subjects
Goodness-of-link test
Goodness-of-link test Local influence Masking Partial influence
Local influence
Masking
Partial influence
Masking
Goodness-of-link test
Local influence
Partial influence
Online AccessGet full text

Cover

More Information
Summary:The evidence for a hypothesized link function in a generalized linear model may depend crucially on one or a few observations. Global and local influence diagnostics are proposed to assess influence on the goodness-of-link test. It is found that the local influence method can reveal influential observations on the link even when masking is present.
ISSN:0167-7152
1879-2103
DOI:10.1016/S0167-7152(96)00049-1