scientific article; zbMATH DE number 1069583

From MaRDI portal
Publication:4356544





There has been considerable development during the last twenty years or so on the question of time series model selection. The main objective of this paper is to review these developments, but restricted to only the class of univariate linear time series models. Although it is not our intention to provide a totally comprehensive survey of the vast literature of this topic, it is hoped that a view of the current state of the art may be gleaned from this review. It will be convenient and useful to discuss first the question of autoregressive model selection, both from a nonparametric and a parametric point of view. The question of ARMA model selection, stressing in particular the computational aspects, is discussed in section 5.3. Further extensions of the model selection procedures, especially to processes with infinite variance and for prediction more than one step ahead, are discussed in section 5.4. For the entire collection see [Zbl 0846.00010].











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4356544)