JSM 2011 Online Program

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

Abstract Details

Activity Number: 171
Type: Contributed
Date/Time: Monday, August 1, 2011 : 10:30 AM to 12:20 PM
Sponsor: Section on Health Policy Statistics
Abstract - #301904
Title: Exact Logistic Regression for Nested Data with with Varying Dispersion Across Clusters
Author(s): Trent Lalonde*+ and Jeffrey Wilson
Companies: University of Northern Colorado and Arizona State University
Address: McKee Hall 520, Greeley, CO, 80634,
Keywords: Binary Response ; Correlated Response ; Logistic Regression ; Exact Methods
Abstract:

Exact methods for binary responses are often necessary when asymptotic methods such as maximum likelihood fail for small sample sizes (Cox, 1970). For the situation of correlated data, logistic methods have been proposed for the asymptotic case (Connolly and Liang, 1988), and also for the exact case (Corcoran et al, 2001). Exact logistic models for data showing multiple levels of clustering have been investigated by Troxler et al, 2011. These methods have shown utility in modeling correlated binary responses, such as repeated observation of success for individuals.

This paper extends a brief discussion of Troxler et al, 2011, of exact logistic models that allow for varying magnitudes of correlation within clusters. This extension is important because it allows the data greater control in determining an appropriate model. The model is presented, and computational considerations are discussed for various cluster and sample sizes. A comparison is made between these extensions and existing exact logistic regression techniques.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2011 program




2011 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.