|
Activity Number:
|
361
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Nonparametric Statistics
|
| Abstract - #302010 |
|
Title:
|
Kernel Reweighting for Inference on Time Series
|
|
Author(s):
|
Kristofer Jennings*+
|
|
Companies:
|
Purdue University
|
|
Address:
|
Department of Statistics, West Lafayette, IN, 47907-2066,
|
|
Keywords:
|
block bootstrap ; tapering ; random weighting bootstrap
|
|
Abstract:
|
Many bootstrap-based methods for time series inference involve the construction of a "pseudo series" of blocks of the original observations. The kernel reweighting procedure uses a correlated reweighting of the original statistic that does not rely on having a reconstructed series. As such, it resembles the tapered block bootstrap of Paparoditis and Politis (2001) without restrictions on the size of tapering window. The asymptotic properties are similar to those of the tapered block bootstrap for stationary series. However, the flexibility of reweighting allows for asymptotic improvement via iteration. A criterion for choosing block size is given. An empirical likelihood formulation is also discussed.
|
- 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 2008 program |
