|
Activity Number:
|
210
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 7, 2006 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #307409 |
|
Title:
|
Model Identification and Forecasting of Stationary Models with GARCH(P,Q) Errors
|
|
Author(s):
|
Melody Ghahramani*+
|
|
Companies:
|
University of Manitoba
|
|
Address:
|
Department of Statistics, Winnipeg, MB, R3T 2N2, Canada
|
|
Keywords:
|
ARMA-GARCH ; kurtosis ; forecasting ; autocorrelation ; RCA
|
|
Abstract:
|
Financial returns are often modeled as autoregressive time series with innovations having conditional heteroscedastic variances, especially with GARCH processes. We have extended the results of Thavaneswaran et al. (2005) on the kurtosis of a zero-mean GARCH process for a class of stationary ARMA (p, q) models with GARCH errors. The autocorrelation of the squared process will be useful in identifying the order of the GARCH processes. Moment properties, including kurtosis of Random Coefficient Autoregressive (RCA) models with GARCH errors are derived. We also study the forecasting problem for stationary ARMA (p, q) financial time series with errors having GARCH errors. Closed form expressions for the l-steps ahead forecasts are also given in terms of weights for stationary ARMA-GARCH models.
|
- 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 2006 program |
