Activity Number:
|
179
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, August 13, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
General Methodology
|
Abstract - #300600 |
Title:
|
On the Asymptotic Distribution of a Multivariate GR-Estimate for a VAR(p) Time Series
|
Author(s):
|
Jeff Terpstra*+ and M. Rao
|
Affiliation(s):
|
North Dakota State University and North Dakota State University
|
Address:
|
P.O. Box 5575 Waldron 201H, Fargo, North Dakota, 58105, USA
|
Keywords:
|
Asymptotic normality ; GR-Estimates ; MGR-Estimates ; Robust ; Vector autoregressive time series
|
Abstract:
|
This presentation discusses a new class of estimates for estimating the parameters of a vector-autoregressive time series. The estimates minimize a sum of weighted pairwise Euclidean distances and extend the univariate GR-estimates of Terpstra et al. (2001a; 2001b) to the multivariate model. Asymptotic linearity properties are derived for the so-called MGR-estimate. Based on these properties, the MGR-estimate is shown to be asymptotically normal at the square root of n rate. This result is valid for both symmetric and asymmetric error distributions. Some asymptotic relative efficiency comparisons between the MGR-estimate and the classical least-squares estimate will also be presented.
|