Edge Deletion Tests in Graphical Models for Multivariate Time Series
Type of content
Conference Contributions - Published
UC permalink
Publisher's DOI/URI
Thesis discipline
Degree name
Publisher
Journal Title
Journal ISSN
Volume Title
Language
Date
2019
Authors
Reale M
Price CJ
Lin S
Ellis R
Abstract
Multivariate autoregressive moving average models can be represented as graphical models with nodes representing either the autoregressive or the moving average components and the edges representing a significant partial correlation among them. Absent edges imply a parsimonious structure of the time series models. Different strategies for testing the presence of edges are presented, assessed and compared.
Description
Citation
Reale M, Price CJ, Lin S, Ellis R (2019). Edge Deletion Tests in Graphical Models for Multivariate Time Series. Denver: Joint Statistical Meeting. 27/07/2019-01/08/2019. JSM Proceedings, Statistical Computing Section. Alexandria, VA: American Statistical Association. 1569-1574.
Keywords
sparsity, graphical lasso, parsimonious models
Ngā upoko tukutuku/Māori subject headings
ANZSRC fields of research
Fields of Research::49 - Mathematical sciences::4905 - Statistics::490511 - Time series and spatial modelling
Rights
All rights reserved unless otherwise stated