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

Abstract Details

Activity Number: 670
Type: Contributed
Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract - #307201
Title: Point and Interval Estimations for Difference of Two Proportions in Incomplete Matched-Paired Data Using Expectation-Maximization Algorithm
Author(s): Adeniyi J. Adewale*+ and Benjamin B. John
Companies: Merck & Co., Inc. and EdgeServe LLC
Address: 351 N. Sumneytown Pike, North Wales, PA, 19454,
Keywords: McNemar's test ; Paired-sample design ; Expectation-Maximization Algorithm ; Dependent proportions
Abstract:

The McNemar's test (1947) is a commonly used score test for comparing two proportions in a paired-sample design. In the spirit of the Miettinen and Nurminen (1985) advocacy for interval estimation methods based on restricted estimation of the variance, the efficient score-based confidence intervals (CIs) for two dependent proportions are derived. Both the McNemar's test and the derived efficient score CIs require complete data - a potential source of bias in the presence of missing data since analysis would be based only on the complete cases. Using the expectation-maximization algorithm a score test and corresponding CIs based on the entire available incomplete data are derived. This test and CIs are equivalent to the original McNemar's test and corresponding CIs with complete data. The properties of the proposed methods are evaluated using simulations.


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.