|
Activity Number:
|
338
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #309853 |
|
Title:
|
Tapered Empirical Likelihood Methods
|
|
Author(s):
|
Daniel Nordman*+
|
|
Companies:
|
Iowa State University
|
|
Address:
|
315 F Snedecor Hall, Ames, IA, 50011,
|
|
Keywords:
|
block ; periodogram ; data taper
|
|
Abstract:
|
The talk addresses two different versions of empirical likelihood (EL) for time series, where both versions involve data tapers. The first formulation involves a ``time domain" EL based on tapered data blocks. Tapering helps to reduce the bias incurred in the studentization steps internal to blockwise EL. Simulation results indicate the data taper improves the coverage accuracy of EL confidence intervals for means. The second tapered version of EL is based on a tapered periodogram for inference on ``frequency domain" parameters, such as autocorrelations. This EL formulation involves no block choices, but the approach is valid for a smaller class of time processes and parameters. Simulation also indicates that tapering increases EL coverage accuracy in the frequency domain.
|
- 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 |
