13.11.2012 Views

Introduction to Categorical Data Analysis

Introduction to Categorical Data Analysis

Introduction to Categorical Data Analysis

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

CHAPTER 7<br />

Loglinear Models for Contingency<br />

Tables<br />

Section 3.3.1 introduced loglinear models as generalized linear models (GLMs) for<br />

count data. One use of them is modeling cell counts in contingency tables. The models<br />

specify how the size of a cell count depends on the levels of the categorical variables<br />

for that cell. They help <strong>to</strong> describe association patterns among a set of categorical<br />

response variables.<br />

Section 7.1 introduces loglinear models. Section 7.2 discusses statistical inference<br />

for model parameters and model checking. When one variable is a binary response<br />

variable, logistic models for that response are equivalent <strong>to</strong> certain loglinear models.<br />

Section 7.3 presents the connection. We shall see that loglinear models are mainly<br />

of use when at least two variables in a contingency table are response variables.<br />

Section 7.4 introduces graphical representations that portray a model’s association<br />

patterns and indicate when conditional odds ratios are identical <strong>to</strong> marginal odds ratios.<br />

The loglinear models of Sections 7.1–7.4 treat all variables as nominal. Section 7.5<br />

presents a loglinear model that describes association between ordinal variables.<br />

7.1 LOGLINEAR MODELS FOR TWO-WAY AND<br />

THREE-WAY TABLES<br />

Consider an I × J contingency table that cross-classifies n subjects. When the<br />

responses are statistically independent, the joint cell probabilities {πij } are determined<br />

by the row and column marginal <strong>to</strong>tals,<br />

πij = πi+π+j , i = 1,...,I, j = 1,...,J<br />

An <strong>Introduction</strong> <strong>to</strong> <strong>Categorical</strong> <strong>Data</strong> <strong>Analysis</strong>, Second Edition. By Alan Agresti<br />

Copyright © 2007 John Wiley & Sons, Inc.<br />

204

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!