Local global neural networks: A new approach for nonlinear time series modeling

Mayte Suárez-Fariñas, Carlos E. Pedreira, Marcelo C. Medeiros

Research output: Contribution to journalArticlepeer-review

Abstract

We propose the local-global neural networks model within the context of time series models. This formulation encompasses some already existing nonlinear models and also admits the mixture of experts approach. We emphasize the linear expert case and extensively discuss the theoretical aspects of the model: stationarity conditions, existence, consistency and asymptotic normality of the parameter estimates, and model identifiability. The proposed model consists of a mixture of stationary and nonstationary linear models and is able to describe "intermittent" dynamics; the system spends a large fraction of time in a bounded region, but sporadically develops an instability that grows exponentially for some time and then suddenly collapses. Intermittency is a commonly observed behavior in ecology and epidemiology, fluid dynamics, and other natural systems. A model-building strategy is also considered, and the parameters are estimated by concentrated maximum likelihood. The procedure is illustrated with two real time series.

Original languageEnglish (US)
Pages (from-to)1092-1107
Number of pages16
JournalJournal of the American Statistical Association
Volume99
Issue number468
DOIs
StatePublished - Dec 2004
Externally publishedYes

Keywords

  • Model building
  • Model identifiability
  • Neural network
  • Nonlinear model
  • Parameter estimation
  • Sunspot number
  • Time series

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Fingerprint

Dive into the research topics of 'Local global neural networks: A new approach for nonlinear time series modeling'. Together they form a unique fingerprint.

Cite this