On the top-Lyapunov exponent of block-triangular stationary random matrices

Author

Gerencser, Laszlo ; Michaletzky, Gyorgy ; Orlovits, Zsanett

Author_Institution

MTA SZTAKI, Comput. & Autom. Inst., Budapest, Hungary

fYear

2007

fDate

2-5 July 2007

Firstpage

5065

Lastpage

5070

Abstract

The objective of this note is to prove a useful, nontrivial technical lemma that is needed in the statistical analysis of linear stochastic systems with random system matrices. This study is motivated by the study of the quasi-maximum-likelihood identification method of certain stochastic volatility processes, called the GARCH (Generalized Autoregressive Conditional Heteroscedasticity) processes.

Keywords

Lyapunov methods; autoregressive processes; linear systems; matrix algebra; maximum likelihood estimation; random processes; statistical analysis; stochastic processes; stochastic systems; GARCH processes; block-triangular stationary random matrices; generalized autoregressive conditional heteroscedasticity; linear stochastic systems; quasimaximum-likelihood identification method; statistical analysis; stochastic volatility processes; top-Lyapunov exponent; Equations; Mathematical model; Random variables; Stochastic systems; Upper bound; Vectors; Writing; GARCH processes; Lyapunov-exponent; almost sure exponential stability; identification; random coefficient linear stochastic systems; random matrix products; stochastic volatility;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference (ECC), 2007 European

Conference_Location

Kos

Print_ISBN

978-3-9524173-8-6

Type

conf

Filename

7068835