Asymptotic behaviour of the sample autocovariance and autocorrelation function of the AR(1) process with ARCH(1) errors
Borkovec, Milan
Bernoulli, Tome 7 (2001) no. 6, p. 847-872 / Harvested from Project Euclid
We consider a stationary AR(1) process with ARCH(1) errors given by the stochastic difference equation ¶ X_t=α X_{t-1}+\sqrt{β+λ X_{t-1}^2}\,ε_{t}\,, t∈{\mathbb{N}}\,, ¶ where the (εt) are independent and identically distributed symmetric random variables. In contrast to ARCH and GARCH processes, AR(1) processes with ARCH(1) errors are not solutions of linear stochastic recurrence equations and there is no obvious way to embed them into such equations. However, we show that they still belong to the class of stationary sequences with regularly varying finite-dimensional distributions and therefore the theory of Davis and Mikosch can be applied. We present a complete analysis of the weak limit behaviour of the sample autocovariance and autocorrelation functions of (Xt), (|Xt|) and (Xt2). The results in this paper can be seen as a natural extension of results for ARCH(1) processes.
Publié le : 2001-12-14
Classification:  ARCH model,  autoregressive process,  extremal index,  geometric ergodicity,  heavy tails,  multivariate regular variation,  point processes,  sample autocovariance function,  strong mixing
@article{1078951126,
     author = {Borkovec, Milan},
     title = {Asymptotic behaviour of the sample autocovariance and autocorrelation function of the AR(1) process with ARCH(1) errors},
     journal = {Bernoulli},
     volume = {7},
     number = {6},
     year = {2001},
     pages = { 847-872},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1078951126}
}
Borkovec, Milan. Asymptotic behaviour of the sample autocovariance and autocorrelation function of the AR(1) process with ARCH(1) errors. Bernoulli, Tome 7 (2001) no. 6, pp.  847-872. http://gdmltest.u-ga.fr/item/1078951126/