The Consequences of Systematic Sampling on Granger Causality

Gulasekaran Rajaguru, Tilak Abeysinghe

Research output: Contribution to conferencePresentationResearchpeer-review

Abstract

In applied econometric literature, the causal inferences are often made based on highly temporally aggregated or systematically sampled data. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference and systematic sampling preserves the direction of causality. This paper examines the issue in detail by plugging in theoretical cross covariances into the limiting values of least squares estimates. The asymptotic distributions of the estimates of systematically sampled process are expressed in terms of the cross covariances of the disaggregated process. An extensive Monte Carlo study is conducted to examine small sample results. Quite contrary to the stationary case where systematic sampling preserves the direction of Granger causality, this paper shows that systematic sampling of integrated series may induce spurious causality, even if, they are used in differenced form. It is observed that in general the most distorting causal inferences are likely at low levels of aggregation where the order of sampling-span just exceeds the actual causal lag. At high levels of aggregation, causal information concentrates in contemporaneous correlations. It is found that if the uni-directional causality runs from a non-stationary series to a stationary or non-stationary series, there is a high likelihood of detecting spurious bi-directional causality
Original languageEnglish
Publication statusPublished - 2004
EventAustralasian Meeting of Econometrics Society - Melbourne, Australia
Duration: 7 Jul 20047 Jul 2004

Conference

ConferenceAustralasian Meeting of Econometrics Society
CountryAustralia
CityMelbourne
Period7/07/047/07/04

Fingerprint

Sampling
Causality
Granger causality
Causal inference
Integrated
Lag
Asymptotic distribution
Small sample
Least squares
Monte Carlo study
Contemporaneous correlation
Temporal aggregation
Applied econometrics

Cite this

Rajaguru, G., & Abeysinghe, T. (2004). The Consequences of Systematic Sampling on Granger Causality. Australasian Meeting of Econometrics Society, Melbourne, Australia.
Rajaguru, Gulasekaran ; Abeysinghe, Tilak. / The Consequences of Systematic Sampling on Granger Causality. Australasian Meeting of Econometrics Society, Melbourne, Australia.
@conference{65e1c4faf6384ba3bf004a7b1023b4b5,
title = "The Consequences of Systematic Sampling on Granger Causality",
abstract = "In applied econometric literature, the causal inferences are often made based on highly temporally aggregated or systematically sampled data. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference and systematic sampling preserves the direction of causality. This paper examines the issue in detail by plugging in theoretical cross covariances into the limiting values of least squares estimates. The asymptotic distributions of the estimates of systematically sampled process are expressed in terms of the cross covariances of the disaggregated process. An extensive Monte Carlo study is conducted to examine small sample results. Quite contrary to the stationary case where systematic sampling preserves the direction of Granger causality, this paper shows that systematic sampling of integrated series may induce spurious causality, even if, they are used in differenced form. It is observed that in general the most distorting causal inferences are likely at low levels of aggregation where the order of sampling-span just exceeds the actual causal lag. At high levels of aggregation, causal information concentrates in contemporaneous correlations. It is found that if the uni-directional causality runs from a non-stationary series to a stationary or non-stationary series, there is a high likelihood of detecting spurious bi-directional causality",
author = "Gulasekaran Rajaguru and Tilak Abeysinghe",
year = "2004",
language = "English",
note = "Australasian Meeting of Econometrics Society ; Conference date: 07-07-2004 Through 07-07-2004",

}

Rajaguru, G & Abeysinghe, T 2004, 'The Consequences of Systematic Sampling on Granger Causality' Australasian Meeting of Econometrics Society, Melbourne, Australia, 7/07/04 - 7/07/04, .

The Consequences of Systematic Sampling on Granger Causality. / Rajaguru, Gulasekaran; Abeysinghe, Tilak.

2004. Australasian Meeting of Econometrics Society, Melbourne, Australia.

Research output: Contribution to conferencePresentationResearchpeer-review

TY - CONF

T1 - The Consequences of Systematic Sampling on Granger Causality

AU - Rajaguru, Gulasekaran

AU - Abeysinghe, Tilak

PY - 2004

Y1 - 2004

N2 - In applied econometric literature, the causal inferences are often made based on highly temporally aggregated or systematically sampled data. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference and systematic sampling preserves the direction of causality. This paper examines the issue in detail by plugging in theoretical cross covariances into the limiting values of least squares estimates. The asymptotic distributions of the estimates of systematically sampled process are expressed in terms of the cross covariances of the disaggregated process. An extensive Monte Carlo study is conducted to examine small sample results. Quite contrary to the stationary case where systematic sampling preserves the direction of Granger causality, this paper shows that systematic sampling of integrated series may induce spurious causality, even if, they are used in differenced form. It is observed that in general the most distorting causal inferences are likely at low levels of aggregation where the order of sampling-span just exceeds the actual causal lag. At high levels of aggregation, causal information concentrates in contemporaneous correlations. It is found that if the uni-directional causality runs from a non-stationary series to a stationary or non-stationary series, there is a high likelihood of detecting spurious bi-directional causality

AB - In applied econometric literature, the causal inferences are often made based on highly temporally aggregated or systematically sampled data. A number of theoretical studies have pointed out that temporal aggregation has distorting effects on causal inference and systematic sampling preserves the direction of causality. This paper examines the issue in detail by plugging in theoretical cross covariances into the limiting values of least squares estimates. The asymptotic distributions of the estimates of systematically sampled process are expressed in terms of the cross covariances of the disaggregated process. An extensive Monte Carlo study is conducted to examine small sample results. Quite contrary to the stationary case where systematic sampling preserves the direction of Granger causality, this paper shows that systematic sampling of integrated series may induce spurious causality, even if, they are used in differenced form. It is observed that in general the most distorting causal inferences are likely at low levels of aggregation where the order of sampling-span just exceeds the actual causal lag. At high levels of aggregation, causal information concentrates in contemporaneous correlations. It is found that if the uni-directional causality runs from a non-stationary series to a stationary or non-stationary series, there is a high likelihood of detecting spurious bi-directional causality

M3 - Presentation

ER -

Rajaguru G, Abeysinghe T. The Consequences of Systematic Sampling on Granger Causality. 2004. Australasian Meeting of Econometrics Society, Melbourne, Australia.