Log-linear modeling

Alexander von Eye, Eun-Young Mun, Patrick Mair

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This article describes log-linear models as special cases of generalized linear models. Specifically, log-linear models use a logarithmic link function. Log-linear models are used to examine joint distributions of categorical variables, dependency relations, and association patterns. Three types of log-linear models are discussed, hierarchical models, nonhierarchical models, and nonstandard models. Emphasis is placed on parameter interpretation. It is demonstrated that parameters are best interpretable when they represent the effects specified in the design matrix of the model. Parameter interpretation is illustrated first for a standard hierarchical model, and then for a nonstandard model that includes structural zeros. In a data example, the relationships among race of defendant, race of victim, and death penalty sentence are examined using a log-linear model with all three two-way interactions. Recent developments in log-linear modeling are discussed.

Original languageEnglish
Pages (from-to)218-223
Number of pages6
JournalWiley Interdisciplinary Reviews: Computational Statistics
Volume4
Issue number2
DOIs
StatePublished - 1 Mar 2012

Fingerprint

Log-linear Models
Modeling
Hierarchical Model
Structural Zeros
Categorical variable
Link Function
Generalized Linear Model
Joint Distribution
Model
Penalty
Standard Model
Logarithmic
Interaction

Keywords

  • Hierarchical models
  • Log-linear modeling
  • Nonhierarchical models
  • Nonstandard models
  • Parameter interpretation

Cite this

von Eye, Alexander ; Mun, Eun-Young ; Mair, Patrick. / Log-linear modeling. In: Wiley Interdisciplinary Reviews: Computational Statistics. 2012 ; Vol. 4, No. 2. pp. 218-223.
@article{7e7e8bae145a4049b61b58221a2f11ae,
title = "Log-linear modeling",
abstract = "This article describes log-linear models as special cases of generalized linear models. Specifically, log-linear models use a logarithmic link function. Log-linear models are used to examine joint distributions of categorical variables, dependency relations, and association patterns. Three types of log-linear models are discussed, hierarchical models, nonhierarchical models, and nonstandard models. Emphasis is placed on parameter interpretation. It is demonstrated that parameters are best interpretable when they represent the effects specified in the design matrix of the model. Parameter interpretation is illustrated first for a standard hierarchical model, and then for a nonstandard model that includes structural zeros. In a data example, the relationships among race of defendant, race of victim, and death penalty sentence are examined using a log-linear model with all three two-way interactions. Recent developments in log-linear modeling are discussed.",
keywords = "Hierarchical models, Log-linear modeling, Nonhierarchical models, Nonstandard models, Parameter interpretation",
author = "{von Eye}, Alexander and Eun-Young Mun and Patrick Mair",
year = "2012",
month = "3",
day = "1",
doi = "10.1002/wics.203",
language = "English",
volume = "4",
pages = "218--223",
journal = "Wiley Interdisciplinary Reviews: Computational Statistics",
issn = "1939-5108",
publisher = "John Wiley and Sons Inc.",
number = "2",

}

Log-linear modeling. / von Eye, Alexander; Mun, Eun-Young; Mair, Patrick.

In: Wiley Interdisciplinary Reviews: Computational Statistics, Vol. 4, No. 2, 01.03.2012, p. 218-223.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Log-linear modeling

AU - von Eye, Alexander

AU - Mun, Eun-Young

AU - Mair, Patrick

PY - 2012/3/1

Y1 - 2012/3/1

N2 - This article describes log-linear models as special cases of generalized linear models. Specifically, log-linear models use a logarithmic link function. Log-linear models are used to examine joint distributions of categorical variables, dependency relations, and association patterns. Three types of log-linear models are discussed, hierarchical models, nonhierarchical models, and nonstandard models. Emphasis is placed on parameter interpretation. It is demonstrated that parameters are best interpretable when they represent the effects specified in the design matrix of the model. Parameter interpretation is illustrated first for a standard hierarchical model, and then for a nonstandard model that includes structural zeros. In a data example, the relationships among race of defendant, race of victim, and death penalty sentence are examined using a log-linear model with all three two-way interactions. Recent developments in log-linear modeling are discussed.

AB - This article describes log-linear models as special cases of generalized linear models. Specifically, log-linear models use a logarithmic link function. Log-linear models are used to examine joint distributions of categorical variables, dependency relations, and association patterns. Three types of log-linear models are discussed, hierarchical models, nonhierarchical models, and nonstandard models. Emphasis is placed on parameter interpretation. It is demonstrated that parameters are best interpretable when they represent the effects specified in the design matrix of the model. Parameter interpretation is illustrated first for a standard hierarchical model, and then for a nonstandard model that includes structural zeros. In a data example, the relationships among race of defendant, race of victim, and death penalty sentence are examined using a log-linear model with all three two-way interactions. Recent developments in log-linear modeling are discussed.

KW - Hierarchical models

KW - Log-linear modeling

KW - Nonhierarchical models

KW - Nonstandard models

KW - Parameter interpretation

UR - http://www.scopus.com/inward/record.url?scp=84856703842&partnerID=8YFLogxK

U2 - 10.1002/wics.203

DO - 10.1002/wics.203

M3 - Article

AN - SCOPUS:84856703842

VL - 4

SP - 218

EP - 223

JO - Wiley Interdisciplinary Reviews: Computational Statistics

JF - Wiley Interdisciplinary Reviews: Computational Statistics

SN - 1939-5108

IS - 2

ER -