Testing for Granger Causality with Mixed Frequency Data

Ghysels, Éric; Hill, Jonathan B.; Motegi, Kaiji

doi:10.2139/ssrn.2465448

Cited by 15 publications

(44 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As has been shown in Bai, Ghysels, and Wright (2013), this may be important for forecasting and as has been shown in Ghysels, Hill, and Motegi (2014), also analysis of Granger causality does not require identifiability of all high frequency parameters.…”

Section: High Frequency Ar Systems and Mixed Frequency Datamentioning

confidence: 93%

Multivariate Ar Systems and Mixed Frequency Data: G-Identifiability and Estimation

et al. 2015

View full text Add to dashboard Cite

This paper is concerned with the problem of identifiability of the parameters of a high frequency multivariate autoregressive model from mixed frequency time series data. We demonstrate identifiability for generic parameter values using the population second moments of the observations. In addition we display a constructive algorithm for the parameter values and establish the continuity of the mapping attaching the high frequency parameters to these population second moments. These structural results are obtained using two alternative tools viz. extended Yule Walker equations and blocking of the output process. The cases of stock and flow variables, as well as of general linear transformations of high frequency data, are treated. Finally, we briefly discuss how our constructive identifiability results can be used for parameter estimation based on the sample second moments.The authors want to thank Hans Havlicek, Vienna University of Technology, and Benedikt Pötscher, University of Vienna, for valuable suggestions concerning the proof of Theorem 2. In addition we thank Peter Phillips, several anonymous referees and handling co-editors for valuable comments.

show abstract

Section: High Frequency Ar Systems and Mixed Frequency Datamentioning

confidence: 93%

Multivariate Ar Systems and Mixed Frequency Data: G-Identifiability and Estimation

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Extensions towards representations of higher dimensional multivariate systems as in Ghysels et al (2015b) can be considered, but are left for further research here. Analyzing Granger causality among more than two variables inherently leads to multihorizon causality (see Lütkepohl, 1993 among others).…”

Section: Notationmentioning

confidence: 99%

“…The latter implies the potential presence of a causal chain: for example, in a trivariate system, X may cause Y through an auxiliary variable Z. To abstract from that scenario, Ghysels et al (2015b) often consider cases, in which high-and low-frequency variables are grouped and causality patterns between these groups, viewed as a bivariate system, are analyzed. They study the presence of a causal chain and multi-horizon causality in a Monte Carlo analysis though.…”

Section: Notationmentioning

confidence: 99%

“…Since the work of Ghysels (2015) for stationary series and the extension of Götz, Hecq, and Urbain (2013) and Ghysels and Miller (2013) for the non-stationary and possibly cointegrated case, we can analyze the link between highand low-frequency series in a VAR system treating all variables as endogenous. Ghysels, Motegi, and Hill (2015b) define causality in such a mixed-frequency VAR and develop a corresponding test statistic. Decent size and power properties of their test, however, are dependent on a relatively small difference in sampling frequencies of the variables involved.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Testing for Granger causality in large mixed-frequency VARs

Götz

Hecq

Smeekes

2016

Journal of Econometrics

View full text Add to dashboard Cite

Reproduction permitted only if source is stated.ISBN 978-3-95729-217-9 (Printversion) Non-technical summary Research QuestionWhen dealing with time series sampled at various frequencies it has become common practice to directly incorporate high-frequency information into the econom(etr)ic model at hand. These specifications were first restricted to the single regression case; with the development of the (stacked) mixed-frequency vector autoregressive (MF-VAR) system (Ghysels, 2015) it is now possible to treat all series similarly and investigate causal effects between them. However, if the difference in frequencies between the series involved is large (as, e.g., in a month/working day scenario), estimation accuracy of the system coefficients is exacerbated, implying the detection of causal effects to be potentially inaccurate. To overcome this issue various parameter reduction techniques are introduced and analyzed. These methods are then evaluated in terms of their ability to detect causality patterns between the series under consideration in the resulting restricted model. ContributionTwo parameter reduction techniques are discussed in detail: three reduced rank regression (RRR) model variants and a Bayesian MF-VAR. Using a Monte Carlo experiment both approaches are compared in terms of their Granger causality testing behavior with an unrestricted VAR, a (time-aggregated) low-frequency VAR and the max-test (Ghysels, Motegi, and Hill, 2015a). To further enhance their finite sample properties we develop and evaluate (whenever possible) two bootstrap variants of these tests. Finally, the methods are applied to U.S. data by investigating channels of causality between the monthly growth rate of the industrial production index (IPI) and daily bipower variation (BV) of the S&P500 index. ResultsWe find that, depending on the direction of causality under consideration, a different set of tests results in the best Granger non-causality testing behavior. For the direction from the high-to the low-frequency series, standard testing within the Bayesian MF-VAR, the max-test and the restricted bootstrap version of the Wald test in two RRR model versions performs best. For the reverse direction, the unrestricted bootstrap variants of the Bonferroni-corrected Wald tests within the unrestricted VAR and the RRR models dominate. As far as our application is concerned, Granger causality from BV to IPIgrowth is clearly supported by the data; evidence for causality in the reverse direction, however, only comes from a subset of tests. Maastricht UniversityAbstract We analyze Granger causality testing in a mixed-frequency VAR, where the difference in sampling frequencies of the variables is large. Given a realistic sample size, the number of high-frequency observations per low-frequency period leads to parameter proliferation problems in case we attempt to estimate the model unrestrictedly. We propose several tests based on reduced rank restrictions, and implement bootstrap versions to account for the uncertainty when estimating fac...

show abstract

“…However, its major limitation is that the physical explanation of probabilities is not straightforward, which is unacceptable by engineers sometimes [1]. Granger causality, a dynamic approach, requires a linear regression model [8,13]. In other words, this method describes a linear causality relationship and needs to assume a linear relation between the process variables.…”

Section: Introductionmentioning

confidence: 99%

Capturing Causality for Fault Diagnosis Based on Multi-Valued Alarm Series Using Transfer Entropy

Wang

Zhang

et al. 2017

Entropy

View full text Add to dashboard Cite

Abstract:Transfer entropy (TE) is a model-free approach based on information theory to capture causality between variables, which has been used for the modeling and monitoring of, and fault diagnosis in, complex industrial processes. It is able to detect the causality between variables without assuming any underlying model, but it is computationally burdensome. To overcome this limitation, a hybrid method of TE and the modified conditional mutual information (CMI) approach is proposed by using generated multi-valued alarm series. In order to obtain a process topology, TE can generate a causal map of all sub-processes and modified CMI can be used to distinguish the direct connectivity from the above-mentioned causal map by using multi-valued alarm series. The effectiveness and accuracy rate of the proposed method are validated by simulated and real industrial cases (the Tennessee-Eastman process) to capture process topology by using multi-valued alarm series.

show abstract

Testing for Granger Causality with Mixed Frequency Data

Cited by 15 publications

References 62 publications

Multivariate Ar Systems and Mixed Frequency Data: G-Identifiability and Estimation

Multivariate Ar Systems and Mixed Frequency Data: G-Identifiability and Estimation

Testing for Granger causality in large mixed-frequency VARs

Capturing Causality for Fault Diagnosis Based on Multi-Valued Alarm Series Using Transfer Entropy

Contact Info

Product

Resources

About