Measures of Dependence for Stable AR(1) Models with Time-Varying Coefficients
- 1 February 2008
- journal article
- research article
- Published by Taylor & Francis in Stochastic Models
- Vol. 24 (1) , 58-70
Abstract
In this paper we study the dependence structure for AR(1) models with symmetric α-stable innovations and time-varying coefficients. In this case the covariance function is not defined and therefore other measures of dependence have to be used. We formulate conditions providing the bounded solutions of such systems and give the form of the solution. Then the covariation and the codifference for stable AR(1) models are studied and the asymptotic relation between these two measures of dependence for the considered models is examined.Keywords
This publication has 10 references indexed in Scilit:
- Bounded solutions for ARMA model with varying coefficientsApplicationes Mathematicae, 2004
- Asymptotic behavior of the covariation and the codifference for arma models with stable innovationsCommunications in Statistics. Stochastic Models, 1997
- Can One See $\alpha$-Stable Variables and Processes?Statistical Science, 1994
- Maximum Likelihood Estimation for Periodic Autoregressive Moving Average ModelsTechnometrics, 1985
- Maximum Likelihood Estimation for Periodic Autoregressive Moving Average ModelsTechnometrics, 1985
- PERIODIC AUTOREGRESSIVE‐MOVING AVERAGE (PARMA) MODELING WITH APPLICATIONS TO WATER RESOURCES1Jawra Journal of the American Water Resources Association, 1985
- An approach to modeling seasonally stationary time seriesJournal of Econometrics, 1979
- Characterization of cyclostationary random signal processesIEEE Transactions on Information Theory, 1975
- A Statistical Analysis of Telephone NoiseBell System Technical Journal, 1974
- Time series with periodic structureBiometrika, 1967