|
Activity Number:
|
221
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 5, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #301856 |
|
Title:
|
Time Series Analysis for PARMA Sequences of Less Than Full Rank
|
|
Author(s):
|
Harry Hurd*+
|
|
Companies:
|
The University of North Carolina at Chapel Hill
|
|
Address:
|
Dept of Medical Statistics, Chapel Hill, NC, 27599,
|
|
Keywords:
|
PARMA ; Periodically Correlated ; Nonstationary ; Time Series
|
|
Abstract:
|
In this paper we consider several aspects of time series analysis for periodic ARMA (PARMA) sequences when the observed sequence is of deficient (innovation) rank. We will begin by discussing the rank of PC sequences and of PARMA sequences in particular, where we address conditions on parameters that produce rank deficiency. Finally, we describe some of the considerations of time series analysis of rank deficient PARMA sequences. Specifically, we address the detection of rank deficiency in the time series and the estimation of parameters by likelihood methods.
|
