|
Activity Number:
|
379
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 9, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #306129 |
|
Title:
|
On Robust Forecasting in Dynamic Vector Time Series Models
|
|
Author(s):
|
Pierre Duchesne*+ and Christian Gagné
|
|
Companies:
|
Université de Montréal and Université de Montréal
|
|
Address:
|
Département de mathématiques et statistique, Montréal, PQ, H3C3J7, Canada
|
|
Keywords:
|
multivariate time series ; exogenous variables ; prediction ; robust estimators ; additive outliers
|
|
Abstract:
|
Robust estimation/prediction in multivariate autoregressive models with exogenous variables (VARX) are considered. The conditional least squares estimators (CLS) are known to be non robust when outliers occur. To obtain robust estimators, the method introduced in Duchesne (2005) and Bou Hamad and Duchesne (2005) is generalized. The asymptotic distribution of the new estimators is studied and from this is obtained in particular the asymptotic covariance matrix of the robust estimators. The occurrence of outliers may invalidate the usual conditional prediction intervals. Consequently, the new robust methodology is used to develop robust conditional prediction intervals. In a simulation study, we investigate the finite sample properties of the robust prediction intervals under several scenarios for the occurrence of the outliers, and the new intervals are compared to non-robust intervals.
|
