The Box-Cox Transformation and Non-Iterative Estimation Methods For Ordinal Log-Linear Models

Eric J. Beh, Thomas B Farver

Research output: Contribution to journalArticle

Abstract

Recently Beh and Farver investigated and evaluated three non-iterative procedures for estimating the linear-by-linear parameter of an ordinal log-linear model. The study demonstrated that these non-iterative techniques provide estimates that are, for most types of contingency tables, statistically indistinguishable from estimates from Newton's unidimensional algorithm. Here we show how two of these techniques are related using the Box-Cox transformation. We also show that by using this transformation, accurate non-iterative estimates are achievable even when a contingency table contains sampling zeros.

Original languageEnglish (US)
Pages (from-to)475-484
Number of pages10
JournalAustralian and New Zealand Journal of Statistics
Volume54
Issue number4
DOIs
StatePublished - Dec 2012

Fingerprint

Box-Cox Transformation
Log-linear Models
Contingency Table
Estimate
Zero
Log-linear models
Box-Cox transformation
Contingency table

Keywords

  • Box-Cox transformation
  • Linear-by-linear association
  • Newton's unidimensional method
  • Ordinal log-linear model

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

The Box-Cox Transformation and Non-Iterative Estimation Methods For Ordinal Log-Linear Models. / Beh, Eric J.; Farver, Thomas B.

In: Australian and New Zealand Journal of Statistics, Vol. 54, No. 4, 12.2012, p. 475-484.

Research output: Contribution to journalArticle

@article{bf450766fe8e4ae49f019aac1caccd7e,
title = "The Box-Cox Transformation and Non-Iterative Estimation Methods For Ordinal Log-Linear Models",
abstract = "Recently Beh and Farver investigated and evaluated three non-iterative procedures for estimating the linear-by-linear parameter of an ordinal log-linear model. The study demonstrated that these non-iterative techniques provide estimates that are, for most types of contingency tables, statistically indistinguishable from estimates from Newton's unidimensional algorithm. Here we show how two of these techniques are related using the Box-Cox transformation. We also show that by using this transformation, accurate non-iterative estimates are achievable even when a contingency table contains sampling zeros.",
keywords = "Box-Cox transformation, Linear-by-linear association, Newton's unidimensional method, Ordinal log-linear model",
author = "Beh, {Eric J.} and Farver, {Thomas B}",
year = "2012",
month = "12",
doi = "10.1111/anzs.12007",
language = "English (US)",
volume = "54",
pages = "475--484",
journal = "Australian and New Zealand Journal of Statistics",
issn = "1369-1473",
publisher = "Wiley-Blackwell",
number = "4",

}

TY - JOUR

T1 - The Box-Cox Transformation and Non-Iterative Estimation Methods For Ordinal Log-Linear Models

AU - Beh, Eric J.

AU - Farver, Thomas B

PY - 2012/12

Y1 - 2012/12

N2 - Recently Beh and Farver investigated and evaluated three non-iterative procedures for estimating the linear-by-linear parameter of an ordinal log-linear model. The study demonstrated that these non-iterative techniques provide estimates that are, for most types of contingency tables, statistically indistinguishable from estimates from Newton's unidimensional algorithm. Here we show how two of these techniques are related using the Box-Cox transformation. We also show that by using this transformation, accurate non-iterative estimates are achievable even when a contingency table contains sampling zeros.

AB - Recently Beh and Farver investigated and evaluated three non-iterative procedures for estimating the linear-by-linear parameter of an ordinal log-linear model. The study demonstrated that these non-iterative techniques provide estimates that are, for most types of contingency tables, statistically indistinguishable from estimates from Newton's unidimensional algorithm. Here we show how two of these techniques are related using the Box-Cox transformation. We also show that by using this transformation, accurate non-iterative estimates are achievable even when a contingency table contains sampling zeros.

KW - Box-Cox transformation

KW - Linear-by-linear association

KW - Newton's unidimensional method

KW - Ordinal log-linear model

UR - http://www.scopus.com/inward/record.url?scp=84872706561&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872706561&partnerID=8YFLogxK

U2 - 10.1111/anzs.12007

DO - 10.1111/anzs.12007

M3 - Article

VL - 54

SP - 475

EP - 484

JO - Australian and New Zealand Journal of Statistics

JF - Australian and New Zealand Journal of Statistics

SN - 1369-1473

IS - 4

ER -