Detecting Co-Movements in Non-Causal Time Series

Gianluca Cubadda, Alain Hecq, Sean Telg

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

This paper introduces the notion of common non-causal features and proposes tools to detect them in multivariate time series models. We argue that the existence of co-movements might not be detected using the conventional stationary vector autoregressive (VAR) model as the common dynamics are present in the non-causal (i.e. forward-looking) component of the series. We show that the presence of a reduced rank structure allows to identify purely causal and non-causal VAR processes of order P>1 even in the Gaussian likelihood framework. Hence, usual test statistics and canonical correlation analysis can be applied, where either lags or leads are used as instruments to determine whether the common features are present in either the backward- or forward-looking dynamics of the series. The proposed definitions of co-movements are also valid for the mixed causal—non-causal VAR, with the exception that a non-Gaussian maximum likelihood estimator is necessary. This means however that one loses the benefits of the simple tools proposed. An empirical analysis on Brent and West Texas Intermediate oil prices illustrates the findings. No short run co-movements are found in a conventional causal VAR, but they are detected when considering a purely non-causal VAR.

Original languageEnglish
Pages (from-to)697-715
Number of pages19
JournalOxford Bulletin of Economics and Statistics
Volume81
Issue number3
DOIs
Publication statusPublished - 1 Jun 2019
Externally publishedYes

Fingerprint

time series
Time series
oil price
Vector Autoregressive Process
Reduced Rank
Vector Autoregressive Model
Canonical Correlation Analysis
Multivariate Time Series
Series
Multivariate Models
statistics
Empirical Analysis
Time Series Models
Maximum Likelihood Estimator
Exception
Test Statistic
Likelihood
Valid
Necessary
Movement

Cite this

Cubadda, Gianluca ; Hecq, Alain ; Telg, Sean. / Detecting Co-Movements in Non-Causal Time Series. In: Oxford Bulletin of Economics and Statistics. 2019 ; Vol. 81, No. 3. pp. 697-715.
@article{d5175761fc2c479087b40d26a8d88dd0,
title = "Detecting Co-Movements in Non-Causal Time Series",
abstract = "This paper introduces the notion of common non-causal features and proposes tools to detect them in multivariate time series models. We argue that the existence of co-movements might not be detected using the conventional stationary vector autoregressive (VAR) model as the common dynamics are present in the non-causal (i.e. forward-looking) component of the series. We show that the presence of a reduced rank structure allows to identify purely causal and non-causal VAR processes of order P>1 even in the Gaussian likelihood framework. Hence, usual test statistics and canonical correlation analysis can be applied, where either lags or leads are used as instruments to determine whether the common features are present in either the backward- or forward-looking dynamics of the series. The proposed definitions of co-movements are also valid for the mixed causal—non-causal VAR, with the exception that a non-Gaussian maximum likelihood estimator is necessary. This means however that one loses the benefits of the simple tools proposed. An empirical analysis on Brent and West Texas Intermediate oil prices illustrates the findings. No short run co-movements are found in a conventional causal VAR, but they are detected when considering a purely non-causal VAR.",
author = "Gianluca Cubadda and Alain Hecq and Sean Telg",
year = "2019",
month = "6",
day = "1",
doi = "10.1111/obes.12281",
language = "English",
volume = "81",
pages = "697--715",
journal = "Oxford Bulletin of Economics and Statistics",
issn = "0305-9049",
publisher = "Wiley-Blackwell",
number = "3",

}

Detecting Co-Movements in Non-Causal Time Series. / Cubadda, Gianluca; Hecq, Alain; Telg, Sean.

In: Oxford Bulletin of Economics and Statistics, Vol. 81, No. 3, 01.06.2019, p. 697-715.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Detecting Co-Movements in Non-Causal Time Series

AU - Cubadda, Gianluca

AU - Hecq, Alain

AU - Telg, Sean

PY - 2019/6/1

Y1 - 2019/6/1

N2 - This paper introduces the notion of common non-causal features and proposes tools to detect them in multivariate time series models. We argue that the existence of co-movements might not be detected using the conventional stationary vector autoregressive (VAR) model as the common dynamics are present in the non-causal (i.e. forward-looking) component of the series. We show that the presence of a reduced rank structure allows to identify purely causal and non-causal VAR processes of order P>1 even in the Gaussian likelihood framework. Hence, usual test statistics and canonical correlation analysis can be applied, where either lags or leads are used as instruments to determine whether the common features are present in either the backward- or forward-looking dynamics of the series. The proposed definitions of co-movements are also valid for the mixed causal—non-causal VAR, with the exception that a non-Gaussian maximum likelihood estimator is necessary. This means however that one loses the benefits of the simple tools proposed. An empirical analysis on Brent and West Texas Intermediate oil prices illustrates the findings. No short run co-movements are found in a conventional causal VAR, but they are detected when considering a purely non-causal VAR.

AB - This paper introduces the notion of common non-causal features and proposes tools to detect them in multivariate time series models. We argue that the existence of co-movements might not be detected using the conventional stationary vector autoregressive (VAR) model as the common dynamics are present in the non-causal (i.e. forward-looking) component of the series. We show that the presence of a reduced rank structure allows to identify purely causal and non-causal VAR processes of order P>1 even in the Gaussian likelihood framework. Hence, usual test statistics and canonical correlation analysis can be applied, where either lags or leads are used as instruments to determine whether the common features are present in either the backward- or forward-looking dynamics of the series. The proposed definitions of co-movements are also valid for the mixed causal—non-causal VAR, with the exception that a non-Gaussian maximum likelihood estimator is necessary. This means however that one loses the benefits of the simple tools proposed. An empirical analysis on Brent and West Texas Intermediate oil prices illustrates the findings. No short run co-movements are found in a conventional causal VAR, but they are detected when considering a purely non-causal VAR.

UR - http://www.scopus.com/inward/record.url?scp=85057118555&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057118555&partnerID=8YFLogxK

U2 - 10.1111/obes.12281

DO - 10.1111/obes.12281

M3 - Article

VL - 81

SP - 697

EP - 715

JO - Oxford Bulletin of Economics and Statistics

JF - Oxford Bulletin of Economics and Statistics

SN - 0305-9049

IS - 3

ER -