On Confidence Intervals for a Single Proportion in the Analysis of CTC Images Data

• Activity Number: 29
• Type: Contributed
• Date/Time: Sunday, August 3, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Krishna Saha*+ and Daniel Miller
• Companies: Central Connecticut State University and Central Connecticut State University
• Keywords: beta-binomial ; clustered binary data ; CTC images data ; confidence interval ; proportion

Abstract

In the estimation of sensitivity and specificity in diagnostic medicine when a patient or the diagnostic unit is considered as a cluster, the confidence interval of a single proportion is frequently used. A number of confidence intervals for a single proportion in the case of non-clustered binary data are proposed in the literature and implemented in standard software packages. However, little attention has been paid to extending such inferences using clustered binary data. In this study, we consider several asymptotic procedures, based on parametric and semiparametric models, to construct the confidence interval for a single proportion based on binary outcome data arising in cluster studies. We compare the performance of the five proposed methods, in terms of coverage and expected lengths, with a Monte Carlo simulation study and we illustrate the methodology with an example from the CTC images study.