A causality-in-variance test and its application to financial market prices

Cheung, Yin‐Wong; Ng, Lilian

doi:10.1016/0304-4076(94)01714-x

Cited by 437 publications

(456 citation statements)

References 19 publications

Supporting

Mentioning

434

Contrasting

Unclassified

Order By: Relevance

“…Next, we also account for heteroscedasticity due to volatility clustering in our data as is evident in Figure 1. To take account of possible conditional heteroscedasticity of unknown form (Cheung and Ng 1996), we employ a popular heteroscedasticity-consistent covariance matrix estimator (HCCME) developed by MacKinnon and White (1985), known in the literature as HC3 estimator, for robustifying the classical linear Granger causality test. An alternative way to improve the performance of the classical Granger causality test in the presence of heteroscedasticity is to use a fixed design wild bootstrap procedure as in Hafner and Herwartz (2009).…”

Section: Empirical Procedures and Resultsmentioning

confidence: 99%

Causality Between Us Economic Policy and Equity Market Uncertainties: Evidence from Linear and Nonlinear Tests

Ajmi

Aye

Balcılar

et al. 2015

Journal of Applied Economics

View full text Add to dashboard Cite

This paper examines the causal relationship between economic policy uncertainty (EPU) and equity market uncertainty (EMU) in the US using linear and nonlinear Granger causality tests. We use daily data on the newly developed indexes by Baker et al. (2013aBaker et al. ( ) covering 1985Baker et al. ( :01:01 to 2013 Results from the linear causality tests indicate strong bidirectional causality. We test for parameters stability, and find strong evidence of short run parameter instability, thus invalidating any conclusion from the full sample linear estimations. Therefore we turn to nonlinear tests. Using Hiemstra and Jones (1994), Diks and Panchenko (2006), and Kyrtsou and Labys (2006) symmetric test, we observe a stronger predictive power from EMU to EPU than from EPU to EMU. (2006) test, we find no evidence of positive predictive power from EPU to EMU. However, we find strong evidence of positive predictive power from EMU to EPU and only weak evidence of negative EPU causing EMU. Performing the causality test using the Sato et al. (2007) time-varying method, we find that the causality between EPU and EMU is not constant over time but rather time-varying. Hence, we implement a sub-sample bootstrap rolling window causality tests to fully account for the existence of structural breaks. Using the intensity plots of the p-values from this, we find evidence that EPU can help predict the movements in EMU only around 1993, 2004 and, 2006. However, we find strong evidence that EMU can help predict the movements in EPU throughout the sample period barring around 1998, 2003 and 2005. Further, the analysis of total effects based on the bootstrap sum of coefficients suggests a positive and stronger causal effect from EMU to EPU but smaller and insignificant causality from EPU to EMU. The implications of these findings for both investors and policy makers are provided.

show abstract

Section: Empirical Procedures and Resultsmentioning

confidence: 99%

Causality Between Us Economic Policy and Equity Market Uncertainties: Evidence from Linear and Nonlinear Tests

Ajmi

Aye

Balcılar

et al. 2015

Journal of Applied Economics

View full text Add to dashboard Cite

show abstract

“…In the literature, testing for causality in variance has been based on the residual cross-correlation function (CCF), as in Cheung and Ng (1996), or by estimating of a multivariate GARCH framework, as in Caporale et al (2002). The methodology developed by Cheung and Ng (1996) (extended by Hong, 2001) is a two-step procedure where the estimation of univariate GARCH models is followed by computation of CCFs of squared standardized residuals.…”

Section: Causality-in-variance Testsmentioning

confidence: 99%

“…The methodology developed by Cheung and Ng (1996) (extended by Hong, 2001) is a two-step procedure where the estimation of univariate GARCH models is followed by computation of CCFs of squared standardized residuals. Applications to the analysis of volatility spillovers include Kanas and Kouretas (2002), Alaganar and Bhar (2003) and Hong (2003) investigating causality in variance between black and parallel currency 10 Laurent and Peters (2002) provide details on the log likelihood functions of multivariate GARCH models.…”

Section: Causality-in-variance Testsmentioning

confidence: 99%

“…Dijk et al (2005) suggest that the Cheung and Ng (1996) test could provide unreliable inference about the (non-) existence of causality in variance if structural changes have not been accounted for. 11 Caporale et al (2002) propose an alternative by estimating a multivariate GARCH framework and then testing for the relevant zero restrictions on the conditional variance parameters.…”

Section: Causality-in-variance Testsmentioning

confidence: 99%

See 1 more Smart Citation

The role of credit in the Great Moderation: A multivariate GARCH approach

Grydaki¹,

Bezemer²

2013

Journal of Banking & Finance

View full text Add to dashboard Cite

During the Great Moderation, financial innovation in the U.S. increased the size and scope of credit flows supporting the growth of wealth. We hypothesize that spending out of wealth came to finance a wider range of GDP components such that it smoothed GDP. Both these trends combined would be consistent with a decrease in the volatility of output. We suggest testable implications in terms of both growth of credit and output and volatility of growth. In a multivariate GARCH framework, we test this view for home mortgages and residential investment. We observe unidirectional causality in variance from total output, residential investment and non-residential output to mortgage lending before, but not during the Great Moderation. These findings are consistent with a role for credit dynamics in explaining the Great Moderation.

show abstract

“…Two main approaches have been followed in the literature. On the one hand is the two-step methodology concentrating on the cross correlation function (CCF) of univariate residual estimates (Cheung and Ng, 1996). On the other hand is the (Quasi) Maximum-Likelihood (QML) method which utilizes a parametric specification of volatility dynamics (Hafner and Herwartz, 2004).…”

mentioning

confidence: 99%

Granger causality with signal-dependent noise

2011

View full text Add to dashboard Cite

It is generally believed that the noise variance in in vivo neuronal data exhibits time-varying volatility, particularly signal-dependent noise. Despite a widely used and powerful tool to detect causal influences in various data sources, Granger causality has not been well tailored for time-varying volatility models. In this technical note, a unified treatment of the causal influences in both mean and variance is naturally proposed on models with signal-dependent noise in both time and frequency domains. The approach is first systematically validated on toy models, and then applied to the physiological data collected from Parkinson patients, where a clear advantage over the classical Granger causality is demonstrated.© 2011 Elsevier Inc. All rights reserved. IntroductionThe past few years have witnessed a significant growth in the application of Granger causality to various research areas, especially to physiological recordings and functional MRI data. Proposed by Granger (1969) and further extended to the frequency domain by Geweke (1982), Granger causality has become a powerful tool to detect causal influences and functional connectivity from the huge amount of temporal data easily accessible nowadays.The classical Granger causal analysis is based on an autoregressive (AR) model (Friston, 2009a), which assumes that the current state of a process can be predicted by a linear function of its previous states plus a white noise or innovation process. A process is said to be the cause of another process if the inclusion of its past information can improve the prediction of the second process, i.e., reduce the variance of the prediction error. In spite of its easy implementation, wide justification and successful applications, some of the working assumptions might be an over-simplification when performing time series analysis in some situations. One particular scenario is the violation of time-invariant noise, that is, the volatility changes over time. As a common phenomenon in financial market series, similar characteristics have been observed in many physiological recordings, such as the data from epilepsy patients, Parkinson patients and so on. Because of the close relationship of many series, it is obvious to conjecture that changes in the volatility of one series may have an impact on the mean activity or volatility of another series and vice versa, which means that causal influences may happen on the second order statistics or there may be feedback between the first and second moments of the series. In these scenarios, a simple AR model obviously cannot capture these effects and a more detailed modeling of the volatility of the time series as well as the corresponding causality inference techniques are desired.The development of specific models for changing volatility, or say, conditionally heteroskedastic data, has long been launched, primarily motivated by the research in the literature of economics. A natural development was boosted by Engle's invention of the autoregressive conditional heteroskedasticity (...

show abstract

A causality-in-variance test and its application to financial market prices

Cited by 437 publications

References 19 publications

Causality Between Us Economic Policy and Equity Market Uncertainties: Evidence from Linear and Nonlinear Tests

Causality Between Us Economic Policy and Equity Market Uncertainties: Evidence from Linear and Nonlinear Tests

The role of credit in the Great Moderation: A multivariate GARCH approach

Granger causality with signal-dependent noise

Contact Info

Product

Resources

About