Temporal aggregation and spurious instantaneous causality in multiple time series models

By J. Breitung, Norman R. Swanson. 2002

104 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Causal Discovery from Temporally Aggregated Time Series (2017, background): “For example, Breitung & Swanson (2002) examined the impact of temporal aggregation on Granger causality in vector autoregressive (VAR) models and found that the results of Granger causal analysis heavily depend on temporal aggregation.”
Granger Causality: A Review and Recent Advances (2021, background): “However, for macroeconomic indicators like GDP, the scale of sampling is generally arbitrary and may not reﬂect the true causal dynamics, leading to confounded Granger and instantaneous causality judgments (Zhou et al., 2014; Breitung and Swanson, 2002).”
Testing for Granger Causality with Mixed Frequency Data (2016, unknown): “The subject has been extensively researched ever since, see notably Lütkepohl (1993), Renault et al. (1998), Marcellino (1999), Breitung and Swanson (2002), McCrorie and Chambers (2006), and Silvestrini and Veredas (2008), among others. http://dx.doi.org/10.1016/j.jeconom.2015.07.007 0304-4076/©…”
Temporal aggregation of multivariate GARCH processes (2004, background): “In temporally aggregated VARMA models, Breitung and Swanson (2002) have investigated the effect of so-called spurious instantaneous causality, as first investigated by Renault and Szafarz (1991) and Renault, Sekkat and Szafarz (1998). This occurs if there is no causality between the disaggregated ti”
Macroeconomics and the Reality of Mixed Frequency Data (2016, background): “We now turn 12 Much has been written about the spurious effects temporal aggregation may have on testing for Granger causality, see e.g. Granger (1980, 1988), Lütkepohl (1993), Granger (1995), Renault et al. (1998), Breitung and Swanson (2002), McCrorie and Chambers (2006), among others.”
Granger causality and the sampling of economic processes (2006, methodology): “In the fixed-interval time aggregation literature, Breitung and Swanson (2002) have recently profitably revived a notion of (apparent) instantaneous causality in a multivariate context, giving sufficient conditions to rule it out as an artefact of temporal aggregation.”
Exchange Rates and Commodity Prices : Measuring Causality at Multiple Horizons (2015, background): “…invariant to aggregation: highfrequency data may reveal patterns which are aggregated away in low-frequency data, and causality in low-frequency data can also be spurious; see Tiao and Wei (1976),Wei (1982, 1990), Marcellino (1999), Breitung and Swanson (2002), and Silvestrini and Veredas (2008).”
Temporal aggregation of univariate and multivariate time series models: a survey (2008, unknown): “Indeed, multiplying both sides of (8) by T (L), p ∏”
Testing for Granger Causality in Large Mixed-Frequency VARs (2016, unknown): “Additionally, aggregating a high-frequency variable may lead to “spurious” (non-)causality in the common low-frequency setup (Breitung and Swanson, 2002), This generic specification nests the two dominating aggregation schemes in the literature, Point-in-Time (A = 0, w0 = 1) and Average sampling (A ”
Discovering Temporal Causal Relations from Subsampled Data (2015, background): “" Various contributions have been made on how temporal aggregation changes the Granger causal relations in the data (Rajaguru & Abeysinghe, 2008; Breitung & Swanson, 2002); for instance , temporal aggregation could cause spurious instantaneous correlations in the time series.”
Causal network reconstruction from time series: From theoretical assumptions to practical estimation. (2018, background): “See Breitung and Swanson (2002) and Barnett and Seth (2015) for a discussion on temporal aggregation in time series models.”
STRUCTURAL DECOMPOSITION ANALYSIS APPLIED TO ENERGY AND EMISSIONS: AGGREGATION ISSUES (2012, background): “See, for example, Rossana and Seater (1995) and Breitung and Swanson (2002) .”
Identifiability and estimation of structural vector autoregressive models for subsampled and mixed-frequency time series. (2017, background): “A causal analysis rooted at a slower time scale than the true causal time scale may both miss true interactions and add spurious ones [3, 4, 5, 6].”