Measures of relative model fit

Alan Agresti, Brian S Caffo

Research output: Contribution to journalArticle

Abstract

Most software for statistical model fitting reports the value of the maximized log likelihood function, but the numerical value is difficult to interpret because of its log scale, its dependence on the sample size, and the possible omission of constants. A sample-size-scaled version of the likelihood function summarizes the model fit. A related index uses the mean of the contributions to the likelihood function. The ratio or difference of either index for two models is a summary measure of relative model fit. We discuss these measures and briefly consider interval estimation for them.

Original languageEnglish (US)
Pages (from-to)127-136
Number of pages10
JournalComputational Statistics and Data Analysis
Volume39
Issue number2
DOIs
StatePublished - Apr 28 2002
Externally publishedYes

Fingerprint

Likelihood Function
Sample Size
Interval Estimation
Model Fitting
Statistical Model
Model
Software
Sample size

Keywords

  • AIC
  • Deviance
  • Discrete data
  • Dissimilarity
  • Likelihood function
  • Non-nested models

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Statistics, Probability and Uncertainty
  • Electrical and Electronic Engineering
  • Computational Mathematics
  • Numerical Analysis
  • Statistics and Probability

Cite this

Measures of relative model fit. / Agresti, Alan; Caffo, Brian S.

In: Computational Statistics and Data Analysis, Vol. 39, No. 2, 28.04.2002, p. 127-136.

Research output: Contribution to journalArticle

Agresti, Alan ; Caffo, Brian S. / Measures of relative model fit. In: Computational Statistics and Data Analysis. 2002 ; Vol. 39, No. 2. pp. 127-136.
@article{eab8c7c20a5c417d8f098f69edd8bca8,
title = "Measures of relative model fit",
abstract = "Most software for statistical model fitting reports the value of the maximized log likelihood function, but the numerical value is difficult to interpret because of its log scale, its dependence on the sample size, and the possible omission of constants. A sample-size-scaled version of the likelihood function summarizes the model fit. A related index uses the mean of the contributions to the likelihood function. The ratio or difference of either index for two models is a summary measure of relative model fit. We discuss these measures and briefly consider interval estimation for them.",
keywords = "AIC, Deviance, Discrete data, Dissimilarity, Likelihood function, Non-nested models",
author = "Alan Agresti and Caffo, {Brian S}",
year = "2002",
month = "4",
day = "28",
doi = "10.1016/S0167-9473(01)00054-8",
language = "English (US)",
volume = "39",
pages = "127--136",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",
number = "2",

}

TY - JOUR

T1 - Measures of relative model fit

AU - Agresti, Alan

AU - Caffo, Brian S

PY - 2002/4/28

Y1 - 2002/4/28

N2 - Most software for statistical model fitting reports the value of the maximized log likelihood function, but the numerical value is difficult to interpret because of its log scale, its dependence on the sample size, and the possible omission of constants. A sample-size-scaled version of the likelihood function summarizes the model fit. A related index uses the mean of the contributions to the likelihood function. The ratio or difference of either index for two models is a summary measure of relative model fit. We discuss these measures and briefly consider interval estimation for them.

AB - Most software for statistical model fitting reports the value of the maximized log likelihood function, but the numerical value is difficult to interpret because of its log scale, its dependence on the sample size, and the possible omission of constants. A sample-size-scaled version of the likelihood function summarizes the model fit. A related index uses the mean of the contributions to the likelihood function. The ratio or difference of either index for two models is a summary measure of relative model fit. We discuss these measures and briefly consider interval estimation for them.

KW - AIC

KW - Deviance

KW - Discrete data

KW - Dissimilarity

KW - Likelihood function

KW - Non-nested models

UR - http://www.scopus.com/inward/record.url?scp=0037188232&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037188232&partnerID=8YFLogxK

U2 - 10.1016/S0167-9473(01)00054-8

DO - 10.1016/S0167-9473(01)00054-8

M3 - Article

AN - SCOPUS:0037188232

VL - 39

SP - 127

EP - 136

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

IS - 2

ER -