Generalized Network Autoregressive Processes and the GNAR package

Marina Knight, Kathryn Leeming, Guy Nason, Matthew Nunes

Research output: Contribution to journalArticlepeer-review

17 Citations (SciVal)


This article introduces the GNAR package, which fits, predicts, and simulates from a powerful new class of generalised network autoregressive processes. Such processes consist of a multivariate time series along with a real, or inferred, network that provides information about inter-variable relationships. The GNAR model relates values of a time series for a given variable and time to earlier values of the same variable and of neighbouring variables, with inclusion
controlled by the network structure. The GNAR package is designed to fit this new model, while working with standard ts objects and the igraph package for ease of use.
Original languageEnglish
Pages (from-to)1-36
Number of pages36
JournalJournal of Statistical Software
Issue number5
Early online date29 Nov 2020
Publication statusPublished - 31 Dec 2020

Bibliographical note

Publisher Copyright:
© 2020, American Statistical Association. All rights reserved.

Copyright 2020 Elsevier B.V., All rights reserved.


  • Missing data
  • Multivariate time series
  • Network time series
  • Networks

ASJC Scopus subject areas

  • Software
  • Statistics and Probability
  • Statistics, Probability and Uncertainty


Dive into the research topics of 'Generalized Network Autoregressive Processes and the GNAR package'. Together they form a unique fingerprint.

Cite this