A test statistic for graphical modelling of multivariate time series

Yasumasa Matsuda

doi:10.1093/biomet/93.2.399

A test statistic for graphical modelling of multivariate time series

Yasumasa Matsuda

ECON - Economics and Management

Research output: Contribution to journal › Article › peer-review

14 Citations (Scopus)

Abstract

A graphical model for multivariate time series is a concept extended by Dahlhaus (2000) from that for a random vector to a multivariate time series. We propose a test statistic for identifying the model based on the Kullback-Leibler divergence between two graphical models. The null distribution is shown to be asymptotically normal with mean and variance which depend just on the dimensions of the graphs.

Original language	English
Pages (from-to)	399-409
Number of pages	11
Journal	Biometrika
Volume	93
Issue number	2
DOIs	https://doi.org/10.1093/biomet/93.2.399
Publication status	Published - 2006 Jun

Keywords

Asymptotic normality
Backward stepwise selection
Conditional independence
Graphical model
Kullback-Liebler divergence
Periodogram
Spectral density matrix
Test statistic

Access to Document

10.1093/biomet/93.2.399

Cite this

@article{1109079539f14d8ead0a7c7dd59a019d,

title = "A test statistic for graphical modelling of multivariate time series",

abstract = "A graphical model for multivariate time series is a concept extended by Dahlhaus (2000) from that for a random vector to a multivariate time series. We propose a test statistic for identifying the model based on the Kullback-Leibler divergence between two graphical models. The null distribution is shown to be asymptotically normal with mean and variance which depend just on the dimensions of the graphs.",

keywords = "Asymptotic normality, Backward stepwise selection, Conditional independence, Graphical model, Kullback-Liebler divergence, Periodogram, Spectral density matrix, Test statistic",

author = "Yasumasa Matsuda",

year = "2006",

month = jun,

doi = "10.1093/biomet/93.2.399",

language = "English",

volume = "93",

pages = "399--409",

journal = "Biometrika",

issn = "0006-3444",

publisher = "Oxford University Press",

number = "2",

}

TY - JOUR

T1 - A test statistic for graphical modelling of multivariate time series

AU - Matsuda, Yasumasa

PY - 2006/6

Y1 - 2006/6

N2 - A graphical model for multivariate time series is a concept extended by Dahlhaus (2000) from that for a random vector to a multivariate time series. We propose a test statistic for identifying the model based on the Kullback-Leibler divergence between two graphical models. The null distribution is shown to be asymptotically normal with mean and variance which depend just on the dimensions of the graphs.

AB - A graphical model for multivariate time series is a concept extended by Dahlhaus (2000) from that for a random vector to a multivariate time series. We propose a test statistic for identifying the model based on the Kullback-Leibler divergence between two graphical models. The null distribution is shown to be asymptotically normal with mean and variance which depend just on the dimensions of the graphs.

KW - Asymptotic normality

KW - Backward stepwise selection

KW - Conditional independence

KW - Graphical model

KW - Kullback-Liebler divergence

KW - Periodogram

KW - Spectral density matrix

KW - Test statistic

UR - http://www.scopus.com/inward/record.url?scp=33745594287&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745594287&partnerID=8YFLogxK

U2 - 10.1093/biomet/93.2.399

DO - 10.1093/biomet/93.2.399

M3 - Article

AN - SCOPUS:33745594287

SN - 0006-3444

VL - 93

SP - 399

EP - 409

JO - Biometrika

JF - Biometrika

IS - 2

ER -

A test statistic for graphical modelling of multivariate time series

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this