Time Series: State Space Methods

Siem J. Koopman*, Jacques J.F. Commandeur

*Corresponding author for this work

Research output: Chapter in Book / Report / Conference proceedingChapterAcademicpeer-review

Abstract

State space modeling provides a unified methodology for treating a wide range of problems in time series analysis. The Kalman filter and its related methods have become key tools in the analysis of time series in economics, finance, and in many other fields as well. In an increasingly more complex world, static and dynamic models have proven to be too limited in empirical and relevant policy studies. The modeling of time-varying features in a time series has been given much attention recently. In this article we review and provide some adequate details and guidance for the adaptation of state space methods in univariate and multivariate time series analysis. We provide more detailed discussions for linear Gaussian model formulations and more concise reviews for nonlinear and non-Gaussian departures.

Original languageEnglish
Title of host publicationInternational Encyclopedia of the Social & Behavioral Sciences: Second Edition
PublisherElsevier Inc.
Pages354-361
Number of pages8
ISBN (Electronic)9780080970875
ISBN (Print)9780080970868
DOIs
Publication statusPublished - 26 Mar 2015

Keywords

  • ARIMA components
  • Dynamic factor model
  • Dynamic linear model
  • Exponential family models
  • Forecasting
  • Kalman filter
  • Latent risk model
  • Missing values
  • Particle filter
  • Poisson counts
  • Seasonal adjustment
  • Signal extraction
  • Simulation
  • Smoothing
  • State space model
  • Stochastic volatility model
  • Structural time series model
  • Unobserved components time series model

Fingerprint Dive into the research topics of 'Time Series: State Space Methods'. Together they form a unique fingerprint.

Cite this