This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 259
Type: Contributed
Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
Sponsor: Health Policy Statistics Section
Abstract - #307136
Title: Comparison of Quadratic Inference Functions and Generalized Estimating Equations: Longitudinal Case Study of Act 1220 of 2003 Health Policy in Arkansas
Author(s): Zoran Bursac*+ and D. Keith Williams and Clinton Heath Gauss and Page Moore and Melanie E. Goodell and Rebecca A. Krukowski and Delia Smith West and Martha M. Phillips and James M. Raczynski
Companies: University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences and University of Arkansas for Medical Sciences
Address: , Little Rock, AR, 72205,
Keywords: GEE ; QIF
Abstract:

Generalized estimating equations (GEE) lack in the area of model selection due to the absence of a goodness-of-fit (GOF) test. Quadratic inference functions (QIF) carry all the strengths of GEE and also produce more robust estimates when the correlation is misspecified, along with a likelihood-ratio-like goodness-of-fit statistic. We compared GEE and QIF through an example of longitudinal policy evaluation. The outcome was compliance with or early adoption of a no access to a beverage vending machines policy. We compared the two methods based on ease of model selection, sensitivity with respect to different working correlation matrices, and efficiency of parameter estimates. QIF allowed for a more detailed assessment of the overall significance along with the fit of the model. QIF produced more efficient parameter estimates than GEE when a less appropriate correlation structure was used.


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 2010 program




2010 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.