JSM 2004 Online Program

Abstract #301959

This is the preliminary program for the 2004 Joint Statistical Meetings in Toronto, Canada. Currently included in this program is the "technical" program, schedule of invited, topic contributed, regular contributed and poster sessions; Continuing Education courses (August 7-10, 2004); and Committee and Business Meetings. This on-line program will be updated frequently to reflect the most current revisions.

To View the Program:
You may choose to view all activities of the program or just parts of it at any one time. All activities are arranged by date and time.

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2004 Program page

Activity Number:	121
Type:	Contributed
Date/Time:	Monday, August 9, 2004 : 12:00 PM to 1:50 PM
Sponsor:	Biopharmaceutical Section
Abstract - #301959
Title:	Exact Analysis of Correlated Binary Data Repeated Measures/RBD Design
Author(s):	Dar Shong Hwang*+ and Chyi-Hung Hsu
Companies:	BRSI and Novartis Pharmaceuticals
Address:	PO Box 562, Florham Park, NJ, 07932,
Keywords:	correlated binary data ; repeated measurement design ; randomized block design ; multinomial distribution ; exact conditional distribution ; optimality
Abstract:	A parametric multinomial distribution was used to describe each of the patients having repeated binary response data. The hypothesis of equal response rates is re-expressed in terms of a subset of multinomial parameters. Existence and optimality of conditional distributions which depend on these parameters were presented. The exact distributions were then derived for the case of three or more time points which appears to be easier to follow. Reduction to the familiar Freeman-Halton (F-H) exact test for the comparison of k independent binomials was proven. Under Ho, the null distribution reduces to that of F-H test when applied to a RBD design. Hence, computation of Type I error and approximation by Mantel-Haenszel (M-H) procedure using existing software is shown. Application to D.R. Cox's exact test of dose-response relationship is shown. Generalization to multicategories of nominal response data per time point resulted in "Generalized" Cochran's Q test, with relationship to existing tests, software, and M-H approximation as in the binary case. The approach is of the population model type, not the randomization model type, having the advantage of inference.

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 2004 program