Generating multivariate categorical variates using the iterative proportional fitting algorithm

Research output: Contribution to journalArticle

Abstract

Two recent papers have suggested methods for generating correlated binary data with fixed marginal distributions and specified degrees of pairwise association. Emrich and Piedmonte suggested a method based on the existence of a multivariate normal distribution, while Lee suggested methods based on linear programming and Archimedian copulas. In this paper, a simpler method is described using the iterative proportional fitting algorithm for generating an n-dimensional distribution of correlated categorical data with specified margins of dimension 1, 2, …, k <n. An example of generating a distribution for a generalized estimating equations (GEE) model is discussed.

Original languageEnglish (US)
Pages (from-to)134-138
Number of pages5
JournalAmerican Statistician
Volume49
Issue number2
DOIs
StatePublished - 1995

Fingerprint

Categorical
Directly proportional
Correlated Binary Data
Correlated Data
Generalized Estimating Equations
Multivariate Normal Distribution
Nominal or categorical data
Copula
Marginal Distribution
Margin
n-dimensional
Linear programming
Pairwise
Model

Keywords

  • Correlated outcomes
  • Generalized estimating equations
  • Loglinear models
  • Random number generation

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Generating multivariate categorical variates using the iterative proportional fitting algorithm. / Gange, Stephen J.

In: American Statistician, Vol. 49, No. 2, 1995, p. 134-138.

Research output: Contribution to journalArticle

@article{0ede65d4e9904d5c96ecac1c2703dc1f,
title = "Generating multivariate categorical variates using the iterative proportional fitting algorithm",
abstract = "Two recent papers have suggested methods for generating correlated binary data with fixed marginal distributions and specified degrees of pairwise association. Emrich and Piedmonte suggested a method based on the existence of a multivariate normal distribution, while Lee suggested methods based on linear programming and Archimedian copulas. In this paper, a simpler method is described using the iterative proportional fitting algorithm for generating an n-dimensional distribution of correlated categorical data with specified margins of dimension 1, 2, …, k",
keywords = "Correlated outcomes, Generalized estimating equations, Loglinear models, Random number generation",
author = "Gange, {Stephen J}",
year = "1995",
doi = "10.1080/00031305.1995.10476130",
language = "English (US)",
volume = "49",
pages = "134--138",
journal = "American Statistician",
issn = "0003-1305",
publisher = "American Statistical Association",
number = "2",

}

TY - JOUR

T1 - Generating multivariate categorical variates using the iterative proportional fitting algorithm

AU - Gange, Stephen J

PY - 1995

Y1 - 1995

N2 - Two recent papers have suggested methods for generating correlated binary data with fixed marginal distributions and specified degrees of pairwise association. Emrich and Piedmonte suggested a method based on the existence of a multivariate normal distribution, while Lee suggested methods based on linear programming and Archimedian copulas. In this paper, a simpler method is described using the iterative proportional fitting algorithm for generating an n-dimensional distribution of correlated categorical data with specified margins of dimension 1, 2, …, k

AB - Two recent papers have suggested methods for generating correlated binary data with fixed marginal distributions and specified degrees of pairwise association. Emrich and Piedmonte suggested a method based on the existence of a multivariate normal distribution, while Lee suggested methods based on linear programming and Archimedian copulas. In this paper, a simpler method is described using the iterative proportional fitting algorithm for generating an n-dimensional distribution of correlated categorical data with specified margins of dimension 1, 2, …, k

KW - Correlated outcomes

KW - Generalized estimating equations

KW - Loglinear models

KW - Random number generation

UR - http://www.scopus.com/inward/record.url?scp=21844498253&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21844498253&partnerID=8YFLogxK

U2 - 10.1080/00031305.1995.10476130

DO - 10.1080/00031305.1995.10476130

M3 - Article

AN - SCOPUS:21844498253

VL - 49

SP - 134

EP - 138

JO - American Statistician

JF - American Statistician

SN - 0003-1305

IS - 2

ER -