|
Activity Number:
|
513
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 10, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #307412 |
|
Title:
|
Rank-Based Estimation for Autoregressive Moving Average Time Series Models
|
|
Author(s):
|
Beth Andrews*+
|
|
Companies:
|
Northwestern University
|
|
Address:
|
Department of Statistics, Evanston, IL, 60208,
|
|
Keywords:
|
time series ; autoregressive moving average models ; rank estimation
|
|
Abstract:
|
A rank-based technique is used to estimate the parameters of autoregressive moving average (ARMA) time series models. The estimators minimize the sum of mean-corrected model residuals weighted by a function of residual rank. They are shown to be consistent and asymptotically normal under very mild conditions on the noise distribution, and so the estimation technique is robust. Because the weight function can be chosen so that rank estimation has the same asymptotic efficiency as maximum likelihood estimation, the estimators are also relatively efficient. The relative efficiency of the estimators extends to the unknown noise distribution case since rank estimation with the Wilcoxon weight function (a linear weight function) is nearly as efficient as maximum likelihood for a large class of noise distributions.
|
