Activity Number:
|
198
- SPEED: Nonparametric Statistics: Estimation, Testing, and Modeling
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 30, 2018 : 11:35 AM to 12:20 PM
|
Sponsor:
|
Section on Nonparametric Statistics
|
Abstract #332708
|
|
Title:
|
Coverage Probability of Empirical Likelihood for Dependent Data
|
Author(s):
|
Guangxing Wang* and Wolfgang Polonik
|
Companies:
|
University of California, Davis and University of California, Davis
|
Keywords:
|
Coverage probibility;
Empirical likelihood confidence region;
High dimensional;
Dependent data;
Convex hull constraint;
Empirical Likelihood
|
Abstract:
|
The upper limit of the coverage probability is a severe limitation on the usefulness of the empirical likelihood confidence region. It is well-known that this upper limit is caused by the convex hull constraint in the construction of the empirical likelihood. Various methods for relaxing the convex hull constraint have been discussed in the literature, but mainly for iid data. In this talk we consider multivariate weakly dependent data, where we present a method that relaxes the convex hull constraint, and still maintains desirable characteristics of the resulting empirical likelihood methods. For example, we will show that the Bartlett corrected coverage error rate can be achieved with this method.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2018 program
|