Order 1 autoregressive process of finite length

Abstract

The stochastic processes of finite length defined by recurrence relations request additional relations specifying the first terms of the process analogously to the initial conditions for the differential equations. As a general rule, in time series theory one analyzes only stochastic processes of infinite length which need no such initial conditions and their properties are less difficult to be determined.

In this paper we compare the properties of the order 1 autoregressive processes of finite and infinite length and we prove that the time series length has an important influence mainly if the serial correlation is significant. These different properties can manifest themselves as transient effects produced when a time series is numerically generated. We show that for an order 1 autoregressive process the transient behavior can be avoided if the first term is a Gaussian random variable with standard deviation equal to that of the theoretical infinite process and not to that of the white noise innovation.

Authors

Călin Vamoş

Ştefan M. Şoltuz

Maria Crăciun

Keywords

Autoregressive process; spectral analysis; time series.

Paper coordinates

C. Vamoş, Ş.M. Şoltuz, M. Crăciun, Order 1 autoregressive process of finite length. Rev. Anal. Numér. Théor. Approx., 36 (2007) 2, 199-214.

References

see the expanding block below.

PDF

About this paper

Journal

Rev. Anal. Numér. Théor. Approx.

Publisher Name

Editura Academiei Romane

Print ISSN

?

Online ISSN

?

2007

Related Posts