The distortionary effects of temporal aggregation on Granger causality

Gulasekaran Rajaguru, Tilak Abeysinghe

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

Abstract

Economists often have to use temporally aggregated data in causality tests. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference. This paper provides a quantitative assessment of the magnitude of the distortions created by temporal aggregation by plugging in theoretical cross covariances into the limiting values of least
squares estimates. Some Monte Carlo results and an application are provided to assess the impact in small samples. It is observed that in general the most distorting causal inferences are likely at low levels of temporal aggregation. At high levels of aggregation, causal information concentrates in contemporaneous correlations. At present, a data-based approach is not available to establish
the direction of causality between contemporaneously correlated variables.
Original languageEnglish
Title of host publicationSome recent developments in statistical theory and applications
Subtitle of host publicationSelected Proceedings of the International Conference on Recent Developments in Statistics, Econometrics and Forecasting, University of Allahabad, India, December 27-28, 2010
EditorsKumar Kuldeep, Anoop Chaturvedi
Place of PublicationBoca Raton, Florida
PublisherBrown Walker Press
Pages38-56
Number of pages19
ISBN (Print)9781612335735
Publication statusPublished - 2012
Event2001 Econometric Society Australasian Meetings - Auckland, New Zealand
Duration: 6 Jul 20019 Jul 2001

Conference

Conference2001 Econometric Society Australasian Meetings
CountryNew Zealand
CityAuckland
Period6/07/019/07/01

Fingerprint

theoretical study
effect
test

Cite this

Rajaguru, G., & Abeysinghe, T. (2012). The distortionary effects of temporal aggregation on Granger causality. In K. Kuldeep, & A. Chaturvedi (Eds.), Some recent developments in statistical theory and applications: Selected Proceedings of the International Conference on Recent Developments in Statistics, Econometrics and Forecasting, University of Allahabad, India, December 27-28, 2010 (pp. 38-56). Boca Raton, Florida: Brown Walker Press.
Rajaguru, Gulasekaran ; Abeysinghe, Tilak. / The distortionary effects of temporal aggregation on Granger causality. Some recent developments in statistical theory and applications: Selected Proceedings of the International Conference on Recent Developments in Statistics, Econometrics and Forecasting, University of Allahabad, India, December 27-28, 2010. editor / Kumar Kuldeep ; Anoop Chaturvedi. Boca Raton, Florida : Brown Walker Press, 2012. pp. 38-56
@inbook{04d25e5c8a2f437fa43e98aab4d24f9d,
title = "The distortionary effects of temporal aggregation on Granger causality",
abstract = "Economists often have to use temporally aggregated data in causality tests. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference. This paper provides a quantitative assessment of the magnitude of the distortions created by temporal aggregation by plugging in theoretical cross covariances into the limiting values of leastsquares estimates. Some Monte Carlo results and an application are provided to assess the impact in small samples. It is observed that in general the most distorting causal inferences are likely at low levels of temporal aggregation. At high levels of aggregation, causal information concentrates in contemporaneous correlations. At present, a data-based approach is not available to establishthe direction of causality between contemporaneously correlated variables.",
author = "Gulasekaran Rajaguru and Tilak Abeysinghe",
year = "2012",
language = "English",
isbn = "9781612335735",
pages = "38--56",
editor = "Kumar Kuldeep and Anoop Chaturvedi",
booktitle = "Some recent developments in statistical theory and applications",
publisher = "Brown Walker Press",

}

Rajaguru, G & Abeysinghe, T 2012, The distortionary effects of temporal aggregation on Granger causality. in K Kuldeep & A Chaturvedi (eds), Some recent developments in statistical theory and applications: Selected Proceedings of the International Conference on Recent Developments in Statistics, Econometrics and Forecasting, University of Allahabad, India, December 27-28, 2010. Brown Walker Press, Boca Raton, Florida, pp. 38-56, 2001 Econometric Society Australasian Meetings, Auckland, New Zealand, 6/07/01.

The distortionary effects of temporal aggregation on Granger causality. / Rajaguru, Gulasekaran; Abeysinghe, Tilak.

Some recent developments in statistical theory and applications: Selected Proceedings of the International Conference on Recent Developments in Statistics, Econometrics and Forecasting, University of Allahabad, India, December 27-28, 2010. ed. / Kumar Kuldeep; Anoop Chaturvedi. Boca Raton, Florida : Brown Walker Press, 2012. p. 38-56.

Research output: Chapter in Book/Report/Conference proceedingChapterResearchpeer-review

TY - CHAP

T1 - The distortionary effects of temporal aggregation on Granger causality

AU - Rajaguru, Gulasekaran

AU - Abeysinghe, Tilak

PY - 2012

Y1 - 2012

N2 - Economists often have to use temporally aggregated data in causality tests. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference. This paper provides a quantitative assessment of the magnitude of the distortions created by temporal aggregation by plugging in theoretical cross covariances into the limiting values of leastsquares estimates. Some Monte Carlo results and an application are provided to assess the impact in small samples. It is observed that in general the most distorting causal inferences are likely at low levels of temporal aggregation. At high levels of aggregation, causal information concentrates in contemporaneous correlations. At present, a data-based approach is not available to establishthe direction of causality between contemporaneously correlated variables.

AB - Economists often have to use temporally aggregated data in causality tests. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference. This paper provides a quantitative assessment of the magnitude of the distortions created by temporal aggregation by plugging in theoretical cross covariances into the limiting values of leastsquares estimates. Some Monte Carlo results and an application are provided to assess the impact in small samples. It is observed that in general the most distorting causal inferences are likely at low levels of temporal aggregation. At high levels of aggregation, causal information concentrates in contemporaneous correlations. At present, a data-based approach is not available to establishthe direction of causality between contemporaneously correlated variables.

UR - http://www.brownwalker.com/book/161233573X

M3 - Chapter

SN - 9781612335735

SP - 38

EP - 56

BT - Some recent developments in statistical theory and applications

A2 - Kuldeep, Kumar

A2 - Chaturvedi, Anoop

PB - Brown Walker Press

CY - Boca Raton, Florida

ER -

Rajaguru G, Abeysinghe T. The distortionary effects of temporal aggregation on Granger causality. In Kuldeep K, Chaturvedi A, editors, Some recent developments in statistical theory and applications: Selected Proceedings of the International Conference on Recent Developments in Statistics, Econometrics and Forecasting, University of Allahabad, India, December 27-28, 2010. Boca Raton, Florida: Brown Walker Press. 2012. p. 38-56