|
Activity Number:
|
676
- Analysis and Reporting: Benefit-Risk and Robust Models
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Medical Devices and Diagnostics
|
|
Abstract #330541
|
Presentation
|
|
Title:
|
Nonparametric Inference for the Coefficient of Variation
|
|
Author(s):
|
Dongliang Wang* and Margaret Formica and Song Liu
|
|
Companies:
|
SUNY Upstate Medical University and SUNY Upstate Medical University and Roswell Park Cancer Institute
|
|
Keywords:
|
Coefficient of variation;
Empirical likelihood;
Jackknife;
Bootstrap;
Confidence interval;
Wilks theorem
|
|
Abstract:
|
The coefficient of variation (CV) is a widely used scaleless measure of variability in many disciplines, but the currently available interval estimators are mainly limited to parametric assumption of normality or standard bootstrapping. Two novel nonparametric methods are proposed via the theory of empirical likelihood, with either newly defined estimating equations or modified jackknife empirical likelihood, in conjunction with bootstrapping for calibration. The proposed methods are illustrated by the analysis of real-life datasets. Extensive empirical studies suggest substantial improvement over the competitors in terms of coverage probabilities of interval estimators.
|
Authors who are presenting talks have a * after their name.
