JSM 2014 Home
Online Program Home
My Program

Abstract Details

Activity Number: 611
Type: Contributed
Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
Sponsor: Section on Medical Devices and Diagnostics
Abstract #313318 View Presentation
Title: Modeling Receiver Operating Characteristics Curve Using Mixture of Skew-T Distriubutions
Author(s): Amay Cheam*+
Companies: University of Guelph
Keywords: Receiver Operating Characteristics Curve ; Mixture of Skew-t Distributions ; Expectation-Maximization Algorithm ; Monte Carlo Simulation ; Binormal Curve
Abstract:

The receiver operating characteristic (ROC) curve is widely applied in measuring the performance of diagnostic tests. In the past, many direct and indirect approaches have been proposed for modelling the ROC curve. Because its ease of manipulation, the Gaussian distribution has been used to model both populations, diseased and non-diseased. In parallel, the finite mixture model has gained popularity as a compelling apparatus in modelling data, especially atypical ones with asymmetric features. Therefore, we propose to model the ROC curve using a mixture of skew-t distributions, leading to a more flexible model that accounts for heterogeneous data, unlike the classical binormal curve. Parameter estimation is performed via the expectation-maximization (EM) algorithm. Furthermore, to circumvent the issue of the absence of a closed-form and to obtain confidence bands for the derived ROC, Monte Carlo simulation procedures are used. We also verify our method through simulation studies and compare our method with existing binormal curve. We present an example involving data from pancreatic cancer that demonstrates that our method is much smoother than some direct approaches.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2014 program




2014 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Professional Development program, please contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

ASA Meetings Department  •  732 North Washington Street, Alexandria, VA 22314  •  (703) 684-1221  •  meetings@amstat.org
Copyright © American Statistical Association.