|
Activity Number:
|
100
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
WNAR
|
| Abstract - #308752 |
|
Title:
|
Noniterative, Semiparametric, Least-Squared Method of ROC Curve Estimation
|
|
Author(s):
|
Xiao-Hua (Andrew) Zhou*+ and Liansheng Tang
|
|
Companies:
|
University of Washington and University of Washington
|
|
Address:
|
1705 Pacific Street NE, Seattle, WA, 98195-7232,
|
|
Keywords:
|
ROC curve ; Semi-parametric ; Biomarkers
|
|
Abstract:
|
The receiver operating characteristic (ROC) curve provides an ideal way to evaluate the discriminating ability of diagnostic tests. In this talk, we introduce a simple estimation method for a semiparametric ROC curves, particularly clustered ROC curve data. Unlike other existing methods, the new approach does not require an iterative algorithm and is easy to implement. Moreover, a nice property of this method is the invariance of estimated parameter vector to any monotone transformation of the measurement scale. We show that the parameter vector in ROC curves is consistent under mild assumptions and derive a consistent estimator of its asymptotic covariance matrix. We derive the theoretical simultaneous confidence bands of estimated ROC curves. The finite sample performance of the proposed procedure is evaluated using Monte Carlo 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 2007 program |
