Online Program Home
My Program

Abstract Details

Activity Number: 120 - SPEED: Nonparametric Statistics: Estimation, Testing, and Modeling
Type: Contributed
Date/Time: Monday, July 30, 2018 : 8:30 AM to 10:20 AM
Sponsor: Section on Nonparametric Statistics
Abstract #329189
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