User: Guest  Login
Document type:
Zeitschriftenaufsatz 
Author(s):
Stelzer, R. 
Title:
Multivariate Markov-switching ARMA processes with regularly varying noise 
Abstract:
The tail behaviour of stationary Rd-valued Markov-Switching ARMA processes driven by a regularly varying noise is analysed. It is shown that under appropriate summability conditions the MS-ARMA process is again regularly varying as a sequence. Moreover, the feasible stationarity condition given in Stelzer (2006) is extended to a criterion for regular variation. Our results complement in particular those of Saporta (2005) where regularly varying tails of one-dimensional MS-AR(1) processes com...    »
 
Keywords:
heavy tails, regular variation, non-linear time series models, stochastic difference equation 
Journal title:
Journal of Multivariate Analysis, 
Year:
2008 
Journal volume:
99 
Journal issue:
Pages contribution:
1177-1190 
Reviewed:
ja 
Language:
en 
Status:
Preprint / submitted 
Semester:
SS 08 
Format:
Text