Functional Time Series

Siegfried Hörmann, Piotr P. Kokoszka

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in Buch/Bericht


This chapter is an account of the recent research that deals with curves observed consecutively over time. The curves are viewed in the framework of functional data analysis, that is, each of them is considered as a whole statistical object. We describe the Hilbert space framework within which the mathematical foundations are developed. We then introduce the most popular model for such data, the functional autoregressive process, and discuss its properties. This is followed by the introduction of a general framework that quantifies the temporal dependence of curves. Within this framework, we discuss analogs of central concepts of time series analysis of scalar data, including the definition and the estimation of an analog of the long-run variance.
TitelTime Series Analysis
UntertitelMethods and Applications
Redakteure/-innenTata Subba Rao
Herausgeber (Verlag)Elsevier B.V.
ISBN (Print)978-0-444-53858-1
PublikationsstatusVeröffentlicht - 2012


NameHandbook of Statistics
ISSN (Print)0169-7161

Fingerprint Untersuchen Sie die Forschungsthemen von „Functional Time Series“. Zusammen bilden sie einen einzigartigen Fingerprint.

  • Dieses zitieren

    Hörmann, S., & Kokoszka, P. P. (2012). Functional Time Series. in T. S. Rao (Hrsg.), Time Series Analysis: Methods and Applications (Band 30, S. 157-186). (Handbook of Statistics; Band 30). Amsterdam: Elsevier B.V..