Leverage and superleverage in nonlinear regression

Roy T. St Laurent, R. Dennis Cook

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

Several measures of the leverage of an observation in a nonlinear regression model are defined and developed. In contrast to the upper bound on the leverage in a linear model, it is found that in a nonlinear model the leverage of an observation may exceed 1. Such a case is said to exhibit superleverage. Relationships between the leverage measures are explored, and several examples are developed to illustrate the proposed methodology.

Original languageEnglish (US)
Pages (from-to)985-990
Number of pages6
JournalJournal of the American Statistical Association
Volume87
Issue number420
DOIs
StatePublished - 1992
Externally publishedYes

Fingerprint

Nonlinear Regression
Leverage
Nonlinear Regression Model
Nonlinear Model
Linear Model
Exceed
Upper bound
Nonlinear regression
Methodology
Observation

Keywords

  • Diagnostic
  • Influential observation
  • Prediction

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Leverage and superleverage in nonlinear regression. / St Laurent, Roy T.; Cook, R. Dennis.

In: Journal of the American Statistical Association, Vol. 87, No. 420, 1992, p. 985-990.

Research output: Contribution to journalArticle

@article{5f37942368e945358ad9840b186992fa,
title = "Leverage and superleverage in nonlinear regression",
abstract = "Several measures of the leverage of an observation in a nonlinear regression model are defined and developed. In contrast to the upper bound on the leverage in a linear model, it is found that in a nonlinear model the leverage of an observation may exceed 1. Such a case is said to exhibit superleverage. Relationships between the leverage measures are explored, and several examples are developed to illustrate the proposed methodology.",
keywords = "Diagnostic, Influential observation, Prediction",
author = "{St Laurent}, {Roy T.} and Cook, {R. Dennis}",
year = "1992",
doi = "10.1080/01621459.1992.10476253",
language = "English (US)",
volume = "87",
pages = "985--990",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "420",

}

TY - JOUR

T1 - Leverage and superleverage in nonlinear regression

AU - St Laurent, Roy T.

AU - Cook, R. Dennis

PY - 1992

Y1 - 1992

N2 - Several measures of the leverage of an observation in a nonlinear regression model are defined and developed. In contrast to the upper bound on the leverage in a linear model, it is found that in a nonlinear model the leverage of an observation may exceed 1. Such a case is said to exhibit superleverage. Relationships between the leverage measures are explored, and several examples are developed to illustrate the proposed methodology.

AB - Several measures of the leverage of an observation in a nonlinear regression model are defined and developed. In contrast to the upper bound on the leverage in a linear model, it is found that in a nonlinear model the leverage of an observation may exceed 1. Such a case is said to exhibit superleverage. Relationships between the leverage measures are explored, and several examples are developed to illustrate the proposed methodology.

KW - Diagnostic

KW - Influential observation

KW - Prediction

UR - http://www.scopus.com/inward/record.url?scp=21144462525&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21144462525&partnerID=8YFLogxK

U2 - 10.1080/01621459.1992.10476253

DO - 10.1080/01621459.1992.10476253

M3 - Article

VL - 87

SP - 985

EP - 990

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 420

ER -