Abstract M09

Comparison of Two Correlated Proportions: Bayesian and Frequentist Approaches
Telba Z. Irony, CDRH - FDA, Carlos A. B. Pereira, University of Sao Paulo, Brazil, Ram C. Tiwari, University of North Carolina

Suppose that a group of individuals is evaluated in order to assess the presence of a certain condition (present = yes, absent = no). After a treatment is administered, these individuals are evaluated again. The objective is to analyze whether or not there was a change - swing - in their state. Due to the longitudinal nature of data, only tests of hypothesis are usually performed - for instance the McNemar test. The estimation of the parameters of interest is usually more complicated. In this poster we suggest Bayesian solutions that make the estimation problem straightforward.


2000 FDA Science Forum | FDA Chapter, Sigma Xi | CFSAN | FDA
Last updated on 2000-FEB-11 by frf