|
Activity Number:
|
275
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
IMS
|
| Abstract - #309403 |
|
Title:
|
Inference for Vector Processes with Long Memory via the EM Algorithm
|
|
Author(s):
|
Nalini Ravishanker*+ and Jeffrey Pai
|
|
Companies:
|
University of Connecticut and University of Manitoba
|
|
Address:
|
215 Glenbrook Road, Storrs, CT, 06269,
|
|
Keywords:
|
EM algorithm ; Financial derivatives ; Long memory ; VARFIMA
|
|
Abstract:
|
A framework for estimation of parameters in vector autoregressive fractionally integrated moving-average (VARFIMA) models via the EM algorithm is presented. It is well known that such processes capture both short-term correlation structure and long-range dependence characteristics of the individual series, as well as interdependence and feedback relationships between the series. Series with Gaussian and heavy-tailed innovations are considered. An application to modeling multivariate daily temperatures at selected measurement sites and its relation to pricing financial derivatives is presented.
|
