Extending the Empirical Likelihood by Domain Expansion

• Activity Number: 347
• Type: Topic Contributed
• Date/Time: Tuesday, August 11, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Nonparametric Statistics
• Author(s): Min Tsao*
• Keywords: Empirical likelihood ; extended empirical likelihood ; coverage accuracy ; Bartlett correction ; similarity transformation ; composite similarity transformation

Abstract

Empirical likelihood confidence regions suffer from an under-coverage problem in that their observed coverage levels tend to be lower than the nominal levels. This is partly due to the convex hull constraint embedded in the formulation of the empirical likelihood, which results in bounded domains and consequently smaller confidence regions. In this talk, I will discuss a geometric approach for expanding the domain of the empirical likelihood which yields an attractive solution to the under-coverage problem.